How to Convert value from double to long type in C ?
Hello Everyone,
I searched on internet but could not find the answer to meet my requirement, I am facing the Conversion problem from double to long without truncation.
In my VC++ application there are some complex tax calculation logic's, and we have to do calculation for all double values and assign those to long type without any truncation.
So, please help me, if having some way for value conversion from double to long in C language ?
Thanking You.
Prat.
How many cents are in a dollar?
You've already lost if you're doing financial calculations in floating point.
Also, since a long is typically 32-bits, and the mantissa of a double is 52 bits, you're always going to lost a lot of bits.
You could use a long long if you wanted, which should be good enough for all the sane currencies in the world.
If you dance barefoot on the broken glass of undefined behaviour, you've got to expect the occasional cut.
If at first you don't succeed, try writing your phone number on the exam paper.
Originally Posted by whiteflags
To expand on what Salem and whiteflags are hinting at; there is no professional financial software in the world that uses any type of floating point calculations to handle transactions. They all deal the the smallest possible representation of the number system in whole amounts. This is due to the inherent inaccuracy of floating point calculations. So, for example if you were to be handling American currency, all your calculations should be done in whole number amounts of cents. So, $5 would be 500 pennies.
Note: This example is still a gross simplification of how financial software actually operates but this should be sufficient enough for what you are trying to do.
EDIT: Read: What every computer scientist should know about floating point arithmetic
Warning: Some or all of my posted code may be non-standard and as such should not be used and in no case looked at.
I wouldn't go that far... I'm betting that right now there are a dozen shiny new Turbo C programmers, still a little moist behind the ears, out there pounding out the best-EVER accounting package just loaded to the brim with floating point math and sequential files loaded into linked lists....
Prat... let me add my vote to those already cast... If you are handling money... use 64bit integers and work in minimum monteary units (usually pennies) right from square 1...
Last edited by CommonTater; 08-21-2011 at 10:47 AM.
For starters, find out whether sizeof(double) == sizeof(long) in VC++.
If it's not then follow what Salem and others have noted in their posts.
LOL......that is probably true. Time to bust out that "Made in India" stamp.
Warning: Some or all of my posted code may be non-standard and as such should not be used and in no case looked at.
