Click here to Skip to main content
15,846,091 members
Articles / Programming Languages / C

Convert Integers to Their Textual Description

Rate me:
Please Sign up or sign in to vote.
3.34/5 (13 votes)
6 May 2014Apache2 min read 157.6K   1.3K   27   52
Convert an integer to its textual description

Sample Image

Preface

Ever wanted a friendlier way to display any given amount to a user? Or, perhaps you're coding a grammatically correct English application, racing game, tracking visitors, or creating an accounting solution. Whatever the case, there are some times when a programmer needs to represent a number as a written word (i.e., 4 becomes "four"). And, that's the goal of this routine.

The original version (as most typical implementations of this) used strings for calculations and conversions and while it worked, it was slower. However, I updated the algorithm to not do this anymore. As a result, I noticed an average 310% speed increase (on a Pentium IV) despite the fact I also added support for much larger numbers as well.

Usage

Using this routine is straightforward. It does make use of C-style strings as they are fast and extremely portable with most environments. The sole routine that does the grunt work is named GetNumWord() and is prototyped as follows...

C++
// prototype/signature
char *GetNumWord (long long llNumber, char *szDest, 
                  unsigned int unLen, bool bOrdinal, bool bUseAnd);

Parameter Descriptions:

llNumber = This is the 64-bit number to convert.
szDest = Pointer to the output buffer (char array).
unLen = Size in bytes of the output buffer (not string length).
bOrdinal = Setting this to true will return the ordinal version of the number (e.g., first); otherwise, the cardinal version is returned (e.g., one).
bUseAnd = This will determine if the and conjunction is used in the output or not (e.g., One Hundred And One).

Example

C++
char szBuffer[100] = {0};
GetNumWord(5001, szBuffer, sizeof(szBuffer), false, false);
GetNumWord(-10, szBuffer, sizeof(szBuffer), false, false);
GetNumWord(123, szBuffer, sizeof(szBuffer), true, true);

Will return...

"Five Thousand One"
"Negative Ten"
"One Hundred And Twenty-Third"

Limitations

The function takes a long long as the number parameter. On most 32-bit and 64-bit systems, this means the lowest number you can pass is <nobr>-9,223,372,036,854,775,808 and the highest number is 9,223,372,036,854,775,807.

Also, 64-bit arithmetic on a 32-bit CPU is a bit slower. However, the difference is negligible for most applications, and 64-bit CPUs are becoming more and more mainstream, so this consideration will soon be obsolete.

Credits & History

<nobr>Article's Author<nobr> - Jeremy Falcon
<nobr>The "and" Conjunction Suggestion<nobr> - benjymous
<nobr>Memory Leak Tip<nobr> - James Curran
  • <nobr>2007-01-13 - Ver. 2.0 released.
  • <nobr>2002-04-12 - Ver. 1.1 released.
  • <nobr>2002-04-01 - Ver. 1.0 released.

License

Permission is granted to anyone to use this software for any purpose on any computer system, and to alter it and redistribute it freely, subject to the following restrictions found in NumWord.c. It's nothing big. Essentially, it states that if your computer blows up then it's not my fault, and if you use this code then give credit where it's due.

License

This article, along with any associated source code and files, is licensed under The Apache License, Version 2.0


Written By
Team Leader
United States United States
I've been in software development since 1994. Over the years I've learned quite a lot in what it takes to complete the process of pushing out a quality product to customers in a timely fashion. As most of my colleagues could attest, there have been many challenges in our new and growing field in the past couple of decades as the industry matures rapidly. Much more so than most others historically speaking.

As such, I've learned one of the best aspects of software engineering is embracing the change that inherently comes along with it as new technologies constantly emerge to help us improve our world one application at a time as we make sense of the overwhelming amount of data now prevalent in the Information Age.

We truly live in a time unlike that ever known to mankind in recorded history, and it is my hope to do my part to help it along to face the challenges and demands of tomorrow.

Comments and Discussions

 
QuestionComments on the code. Pin
Bill_Hallahan14-May-14 16:40
Bill_Hallahan14-May-14 16:40 
AnswerRe: Comments on the code. Pin
Jeremy Falcon14-May-14 17:05
professionalJeremy Falcon14-May-14 17:05 
GeneralRe: Comments on the code. Pin
Bill_Hallahan15-May-14 17:24
Bill_Hallahan15-May-14 17:24 
GeneralRe: Comments on the code. Pin
Jeremy Falcon16-May-14 4:56
professionalJeremy Falcon16-May-14 4:56 
GeneralRe: Comments on the code. Pin
Bill_Hallahan16-May-14 23:34
Bill_Hallahan16-May-14 23:34 
Jeremy Falcon wrote:
Bill_Hallahan wrote:
That wasn't my point. The point was that the code outputs a wrong string value for the value 0x8000000000000000. This is a bug.


You might want to read up on "how to talk to people" and "sarcasm". I'm fully aware of what the point was. Wink | ;)

 

You might want to read up on communication - quoting an irrelevant partial sentence and then stating that it is correct doesn't come across as sarcasm in a text forum where someone has reported a bug to you.

I do realize there is no bug now, and that the magnitude is correct, but not because of your post. It just made no sense. I looked at the code again with fresh eyes today after a good nights sleep. Then I realize why you posted the way you did. Before that, frankly, I thought you were confused. I would rather you would have been straight with me and let me know that I was confused. Unlike some people here, I admit when I make mistakes.
Jeremy Falcon wrote:
Bill_Hallahan wrote:
With a fast divide instruction, I'm not sure if Knuth's algorithm is worth it. I already have python code that converts a number to a text string. (I have this because I once worked with text-to-speech). That also uses the simple divide by 10 method. I'm going to code that in C and time it.


 


I am really surprised that using strtoull and then subtracting '0' (48) from each ASCII digit and using that as an index isn't faster than the code you wrote. Again, the code is now calculating each base10 digit, it just uses the digit as an index and then throws it away. I would expect the C routine to use optimal algorithms and take advantage of Intel parallel instructions (SSE1 and SSE2 instructions) that a C compiler does not generate.


I'm starting to think you haven't even looked at the code or article. It does not take a string and produce a number. Nor does it take a number and produce a string. It takes a number and produces a textual representation of a that number. Much like how you'd write out an amount on a check. What good would strtoull do me if my starting input is an unsigned long long to begin with?

I looked at the code before posting. I named the wrong function, I meant a C function such as sprintf, that can convert a number to base 10. Then the base 10 digits can be converted to indices by subtracting '0' (48). This is the simplest and easiest to read implementation, even if not particularly fast. However, without timing it, I wouldn't assume it isn't fast. You apparently timed an implementation with "strings" because you posted about that in the article. I wonder what implementation with "strings" you timed.

Note, the algorithm I mentioned from the Knuth book can be done with any size integer data type, it need not use type char to store each digit, it an store them in an array of type int.
GeneralRe: Comments on the code. Pin
Jeremy Falcon16-May-14 5:01
professionalJeremy Falcon16-May-14 5:01 
GeneralRe: Comments on the code. Pin
Bill_Hallahan16-May-14 23:21
Bill_Hallahan16-May-14 23:21 
GeneralRe: Comments on the code. Pin
Jeremy Falcon16-May-14 6:31
professionalJeremy Falcon16-May-14 6:31 
GeneralRe: Comments on the code. Pin
Bill_Hallahan17-May-14 0:17
Bill_Hallahan17-May-14 0:17 
GeneralRe: Comments on the code. Pin
Jeremy Falcon17-May-14 4:53
professionalJeremy Falcon17-May-14 4:53 
GeneralRe: Comments on the code. Pin
Bill_Hallahan17-May-14 10:23
Bill_Hallahan17-May-14 10:23 
GeneralRe: Comments on the code. Pin
Jeremy Falcon18-May-14 5:15
professionalJeremy Falcon18-May-14 5:15 
GeneralRe: Comments on the code. Pin
Bill_Hallahan25-May-14 20:31
Bill_Hallahan25-May-14 20:31 
QuestionAn Actor that write code? Pin
Member 761463224-Nov-13 20:00
Member 761463224-Nov-13 20:00 
AnswerRe: An Actor that write code? Pin
Jeremy Falcon17-Apr-14 8:57
professionalJeremy Falcon17-Apr-14 8:57 
GeneralFavor for a favor... some good alternatives. Pin
Shaun Harrington11-Aug-06 3:59
Shaun Harrington11-Aug-06 3:59 
GeneralRe: Favor for a favor... some good alternatives. Pin
Jeremy Falcon13-Jan-07 9:29
professionalJeremy Falcon13-Jan-07 9:29 
Questionhow to represent 1000.25 ???? Pin
Renjith Ramachandran10-Jan-05 19:20
Renjith Ramachandran10-Jan-05 19:20 
AnswerRe: how to represent 1000.25 ???? Pin
Jeremy Falcon13-Jan-07 8:29
professionalJeremy Falcon13-Jan-07 8:29 
QuestionCan we have the PHP? Pin
ChinaHorse18-Mar-04 18:19
ChinaHorse18-Mar-04 18:19 
AnswerRe: Can we have the PHP? Pin
Jeremy Falcon19-Mar-04 10:49
professionalJeremy Falcon19-Mar-04 10:49 
GeneralRe: Can we have the PHP? Pin
ChinaHorse21-Mar-04 1:28
ChinaHorse21-Mar-04 1:28 
GeneralRe: Can we have the PHP? Pin
Jeremy Falcon21-Mar-04 4:25
professionalJeremy Falcon21-Mar-04 4:25 
GeneralYes, SSH Pin
ChinaHorse21-Mar-04 23:17
ChinaHorse21-Mar-04 23:17 
GeneralRe: Yes, SSH Pin
Jeremy Falcon23-Jun-04 9:07
professionalJeremy Falcon23-Jun-04 9:07 

General General    News News    Suggestion Suggestion    Question Question    Bug Bug    Answer Answer    Joke Joke    Praise Praise    Rant Rant    Admin Admin   

Use Ctrl+Left/Right to switch messages, Ctrl+Up/Down to switch threads, Ctrl+Shift+Left/Right to switch pages.