Unexpected exception thrown by Convert.ToDecimal

Question

This is probably an old one but I can't find anything on it. Does anyone have any idea why this:

Convert.ToDecimal("3.14521963414679E-08")

throws

FormatException ("Input string was not in a correct format.")

However this works as expected?

Convert.ToDouble("3.14521963414679E-08")

I thought Convert.ToDecimal could handle exponentials - maybe I got that wrong.

A FormatException: "Input string was not in a correct format." I thought Convert.ToDecimal could handle exponentials? — John Reilly, Jul 11 '13 at 11:25
[`double.Parse()`](http://msdn.microsoft.com/en-us/library/fd84bdyt.aspx) can, but [`decimal.Parse()`](http://msdn.microsoft.com/en-us/library/cafs243z.aspx) can't. Those are the methods called, when doing `Convert.To...`. — Nolonar, Jul 11 '13 at 11:27

Daniel Hilgarth · Accepted Answer · 2013-07-11T11:36:22.820

9

Convert.ToDecimal doesn't support the scientific notation.

It is documented that Convert.ToDecimal internally uses Decimal.Parse and the documentation for Decimal.Parse states that this uses NumberStyles.Number and thus only the following is a valid input:

[ws][sign][digits,]digits[.fractional-digits][ws]

To support scientific notation you will have to use another overload of Decimal.Parse that allows you to specify the NumberStyles to be used:

var result = decimal.Parse("3.14521963414679E-08",
                           NumberStyles.Number | NumberStyles.AllowExponent);

edited Jul 11 '13 at 11:36

answered Jul 11 '13 at 11:25

Daniel Hilgarth

171,043
40
335
443

2

Erh..I don't understand why this answer wasn't selected. – Soner Gönül Jul 11 '13 at 11:37
This actually answered the question even more usefully so I've switched this to be the accepted answer. – John Reilly Jul 11 '13 at 12:00

Soner Gönül · Answer 2 · 2013-07-11T11:52:22.533

Convert.ToDecimal() method uses Decimal.Parse() explicitly.

From MSDN;

Indicates that the numeric string can be in exponential notation. The AllowExponent flag allows the parsed string to contain an exponent that begins with the "E" or "e" character and that is followed by an optional positive or negative sign and an integer. In other words, it successfully parses strings in the form nnnExx, nnnE+xx, and nnnE-xx. It does not allow a decimal separator or sign in the significand or mantissa; to allow these elements in the string to be parsed, use the AllowDecimalPoint and AllowLeadingSign flags, or use a composite style that includes these individual flags.

You can use Decimal.Parse Method (String, NumberStyles) overload of this method which allows you to use NumberStyles enumration like;

Decimal.Parse("3.14521963414679E-08",
              NumberStyles.AllowExponent | NumberStyles.AllowDecimalPoint);

Here a DEMO.

Thanks - that's a really useful addition Soner – John Reilly Jul 11 '13 at 11:45 — John Reilly, Jul 11 '13 at 11:45

score 0 · Answer 3 · answered Jul 11 '13 at 11:31

Because of this->

ToDecimal(String)--Converts the specified string representation of a number to an equivalent decimal number..

ToDouble(String)--Converts the specified string representation of a number to an equivalent double-precision floating-point number.

for further clarification visit- http://msdn.microsoft.com/en-us/library/System.Convert_methods.aspx

Hope it helps..:)

Unexpected exception thrown by Convert.ToDecimal

3 Answers3