Return, by value, a datum having the specified double value. When value is NaN this method guarantees only that a NaN value is stored. The sign and NaN payload bits of a NaN value later retrieved by the theDouble method are unspecified (see also {Special Floating Point Values}.