Converting BigDecimal.toPlainString() to double for performing arithmetic operations

Question

I am getting the following value from my database when I do resultSet.getdouble()

2.0006362409E9

I want to perform arithmetic operations on this number. I tried to get the actual value using BigDecimal.toPlainString(), then I am getting the following value

2000636240.900000095367431640625

When I try to convert BigDecimal.toPlainString() to double I am getting values again in scientific notation.

I want to divide this number -how can I solve this problem?

Why does the scientific notation bother you? You can perform arithmetic directly either on `double` or on `BigDecimal`. The notation is just a print thing. — RealSkeptic, Jul 10 '17 at 12:07
@SangamBelose Maybe you explain why you think *rounding* matters here. — GhostCat, Jul 10 '17 at 12:23
@GhostCat I just want to say that you can do all the calculations using BigDecimal as well. I mentioned RoundingMode because many times system dont require the too large decimal part. — Sangam Belose, Jul 10 '17 at 12:26
@SangamBelose Well, without further explanations, that wasn't really clear. — GhostCat, Jul 10 '17 at 12:26

GhostCat · Answer 1 · 2017-07-10T12:14:24.003

You don't necessarily have to convert to double in order to perform computations.

The wohle point of using BigDecimal is that it allows to deal with arbitrary large numbers regarding all aspects that numbers "need". Meaning: BigDecimal has various methods for doing computations, such as divide().

In other words: unless you are 100% sure that the numbers coming from that database fit into the range for Java double values - you might simply "stick" with using BigDecimal objects. Any computation that is possible with primitive doubles can also be done using the methods provided by the BigDecimal class.

And in case your numbers are in the double range (your example indicates so) - simply use Double.valueOf(String) in order to turn that string into a Double/double. Then you would not need that BigDecimal "detour" in the first place!

I guess my very first version wasn't really too good. And I have to admit: I was too lazy to first check out if the example needs BigDecimal or could go as normal double. And you have nice, precise answer there, too. — GhostCat, Jul 10 '17 at 12:15

score 4 · Answer 2 · answered Jul 10 '17 at 12:11

You have a misconception about internal number representation and user display.

To do calculations with large numbers like yours the BigDecimal class it the right choice. Yes, you cannot use primitive operators (+/-...) and have to use the methods provided by the BigDecimal class (add(),subtract()...). But in return you get almost infinite accuracy while calculating with primitive double values has inherent failure because of fixed number length, especially when dealing with fractions.

Converting BigDecimal.toPlainString() to double for performing arithmetic operations

2 Answers2