Get string from large double value(C#)

Question

Can't find simple way to convert double to string. I need to convert large numbers without distortion. Such as:

double d = 11111111111111111111;
string s = d.ToString();
Console.WriteLine(s);
//1.11111111111111E+19

How to get string value from double value exactly the same as user enter.

11111111111111111111111 => "11111111111111111111111"

1.111111111111111111111 => "1.111111111111111111111"

Any ideas how it can be done?

If you want a value that is [exactly](http://stackoverflow.com/q/588004/11683) what the user entered, `double` is not your choice. — GSerg, Nov 02 '16 at 09:59
Possible duplicate of [Convert double to string](http://stackoverflow.com/questions/533931/convert-double-to-string) — active92, Nov 02 '16 at 09:59
Possible duplicate of [Double to string conversion without scientific notation](http://stackoverflow.com/questions/1546113/double-to-string-conversion-without-scientific-notation) — GSerg, Nov 02 '16 at 10:04
have a look at [this dotnetfiddle](https://dotnetfiddle.net/wjfqeo) — mmushtaq, Nov 02 '16 at 10:07
Having tested, this is very interesting - the number is so large that the compiler (at least in .NET Fiddle) will turn it into scientific notation in memory, causing even a literal string version to turn into `11111111111111100000` — Alfie Goodacre, Nov 02 '16 at 10:08
@AlfieGoodacre This is how floating point numbers are stored in principle. It is not about how big the number is. Please see http://stackoverflow.com/q/588004/11683. — GSerg, Nov 02 '16 at 10:13
In your example, you're not using any decimals. If you only need to support whole numbers, you might want to consider using [BigInteger](https://msdn.microsoft.com/en-us/library/system.numerics.biginteger%28v=vs.110%29.aspx?f=255&MSPPError=-2147217396) — Matthew Watson, Nov 02 '16 at 10:33

score 8 · Accepted Answer · edited May 23 '17 at 12:30

8

double is a floating point type. So it has a limited accuracy. In your example, you could do something like this:

double d = 11111111111111111111;
string s = d.ToString("F0");
Console.WriteLine(s);

But as you'll see,this would output 11111111111111100000 instead of 11111111111111111111,so it has lost accuracy in the process. So the answer here is use the right type for the work. If you need a string, use a string variable to store the value.

Edit

This was the question i was trying to find that explains the problem with floating point math., thanks to @GSerg

edited May 23 '17 at 12:30

Community

1
1

answered Nov 02 '16 at 10:08

Pikoh

7,582
28
53

Which type I can use instead? – Artem Kyba Nov 02 '16 at 10:10
1

That depends on what you are trying to achive. You have not told us anything about the process. – Pikoh Nov 02 '16 at 10:10
Thanks for your answer. Now I realized what the problem is. The double has only fifteen digits and all the rest its trash or zeros or exponentiation. – Artem Kyba Nov 02 '16 at 10:25

Tinwor · Answer 2 · 2016-11-02T10:26:10.750

2

First of all: 11111111111111111111111 is to large for a double value and also this value: 1.111111111111111111111 since the double max decimal length is 17.

By default, a Double value contains 15 decimal digits of precision, although a maximum of 17 digits is maintained internally.

For this reason you should use BigInteger and then ToString for formatting the output.
There is also a library in the nuget Directory called BigRational, never used and seems in Beta stage but probably will help in solving this problem.

edited Nov 02 '16 at 10:26

answered Nov 02 '16 at 10:10

Tinwor

7,765
6
35
56

What about double values? – Artem Kyba Nov 02 '16 at 10:11
Unfortunately there is no equivalent of BigDouble in C# so you must manage yourself the decimal part. – Tinwor Nov 02 '16 at 10:15

Dmitry Bychenko · Answer 3 · 2016-11-02T10:31:33.233

In general case, you can't do this: user can well input, say 123, in many a way:

123
123.00
1.23e2
12.3E1
123.0e+00
1230e-1

etc. When you convert the user input into double you loose the initial format:

string userInput = ...

// double is just 123.0 whatever input has been
double value = double.Parse(userInput);

In case you want to drop exponent if it's possible you can

double value = 11111111111111111111;

string result = value.ToString("#######################");

And, please, notice, that double has 64 bit to store the value, that's why a distortion is inevitable for large numbers:

// possible double, which will be rounded up
double big = 123456789123456789123456789.0;

// 1.2345678912345679E+26
Console.WriteLine(big.ToString("R"));
// 123456789123457000000000000
Console.WriteLine(big.ToString("###########################"));

May be you want BigInteger instead of double:

using System.Numerics;

...

BigInteger value = BigInteger.Parse("111111111111111111111111111111111"); 

// 111111111111111111111111111111111
Console.WriteLine(value.ToString());

Get string from large double value(C#)

3 Answers3