How could I get acceptable precision when converting a float to int gives an erroneous result?

Question

I\'m currently running through a strange case : I have a double value with one decimal, ie. 24.9

from this value I want to get the entire part ie. entire = 24 and, as