Can someone explain or provide a reference to a treatment of the well known Dirac delta function which places it on a better foundation (not too rigorous!) than a switching of orders of integration or of limit taking?

An expression of the delta function that particularly interests me is the Fourier transform of the function f(x) = 1.