Casting from Decimal to Float

Question

I am doing a little bit of mathematical calculations. Here is my code:

   decimal FirstYr = decimal.Round((first / second), 5);

If I am passing first = 20 and second = 34, I am getting the value of 0.58824 in FirstYr. Which is fine. Now I am doing this calculation here in my LINQ:

   ev.HiComm = (float)(FirstYr * 100);

HiComm is float in the DB and this above calculation is passing value: 58.824001312255859

Which I dont want it. I want in my ev.HiComm = 58.824

What I am doing wrong here?

This may interest you: http://stackoverflow.com/questions/122523/why-is-a-sql-float-different-from-a-c-float — womp, Apr 22 '11 at 17:02
Any reason you don't just multiple first line of code by 100 — TheLukeMcCarthy, Apr 22 '11 at 17:06
I can do that, but either way I need to convert it to float! — RG-3, Apr 22 '11 at 17:08
I created a unit test for the code you have above, and it works for me. Hopefully someone will give a real answer, I don't see why moving the * 100 should make it work :-/ — TheLukeMcCarthy, Apr 22 '11 at 18:11

score 3 · Answer 1 · answered Apr 22 '11 at 16:54

3

if you want exactly 3 places after the point, you'll have to change your sql field to be decimal(x,3) where x is 3 plus the maximum places you'll need before the decimal. float is incapable of storing the exact value you want.

answered Apr 22 '11 at 16:54

lincolnk

11,218
4
40
61

process the value read from the db back into your desired format? if your value needs to be that precise, `float` is going to be insufficient. – lincolnk Apr 22 '11 at 16:58
@Mayank Raj: The only thing you can do is round the value back to 3 decimals whenever you fetch it from the DB. – Michael Borgwardt Apr 22 '11 at 16:59
2

@lincolnk: Lack of precision is not the problem. The problem is wanting to store decimal fractions in a binary format. – Michael Borgwardt Apr 22 '11 at 17:01
@Michael: It actually is for creating reports/jobs in MSSQL 2008. I cannot do that. The only option I have is to either change the datatype in DB or change something in my code behind... – RG-3 Apr 22 '11 at 17:07
@Mayank Raj: then change the DB datatype - that's the actually correct solution. – Michael Borgwardt Apr 22 '11 at 17:14

score 0 · Accepted Answer · answered Apr 22 '11 at 17:48

0

It worked like this in the code behind:

 decimal FirstYr = decimal.Round((first / second), 5)*100;
 ev.HiComm = (double)FirstYr ;

Thanks for all your suggestions!

answered Apr 22 '11 at 17:48

RG-3

6,088
19
69
125

1

All you did was cast a decimal to a double which begs the question the reason you are using a float ( decmial ) value in the database to begin with. – Security Hound Apr 22 '11 at 17:51

Casting from Decimal to Float

2 Answers2