Print decimal notation differences under linux and windows

[Bece]

Hey, Hope this question is cool for this forum.
I have a large model that i'm testing under linux and windows. To check the output files i want to simply be able to do a file compare.

Unfortunately printf("%e\n", value), where value = 1.0 results in different output formats under linux and windows.
Windows produces 1.000000e+000
and linux produces 1.000000e+00

As you can see windows displays three values after the e while linux displays 2.

Does anyone know how i can control this?

yeah i could write a hack to convert the output files before comparing them but it would be cool to not have to do that.

Thanks for any help in advance.

 

[cpjust]

I think it should work if you specify the precision to use:

http://www.cppreference.com/stdio/printf.html

ex.

printf( "%[COLOR=red]12.2[/color]e\n", value );

 

[Salem]

That's not all you have to deal with.

http://en.wikipedia.org/wiki/Floating_point

Floats are approximations, and you're likely to find all sorts of very minor differences which are going to result in differences in the output.

--
If you dance barefoot on the broken glass of undefined behaviour, you've got to expect the occasional cut.