I think there are many situations where java is used in development work. I think that Big Decimal may be used when building calculation logic. Suddenly, it's a different type from a normal integer, so I was confused, so I decided to leave it in the article.
https://docs.oracle.com/javase/jp/8/docs/api/java/math/BigDecimal.html> From oracle.docs No, I don't really understand even if it is said mathematically. .. .. So, in other words ... In the case of calculations where floats and doubles cause errors, unintended results may be obtained. For example
double.java
double result = 10d / 6d;
System.out.println(result)
//Output 1.6666666666666666666666666666666666666666
It's like that. As you may know, if you want to calculate "bank rounding", the above example is too irregular and you will be in trouble. That's where Big Decimal comes in. It is often used for so-called core banking calculations.
BigDecimal is different from the normal number calculation and calculates as follows.
BigDecimalCalc.java
//When dividing 1000 of BigDecimal type by 100 of int type
BigDecimal bigNum = BigDecimal.valueOf(1000);
BigDecimal result = bigNum.divide(BigDecimal.valueOf(100), 2, RoundingMode.DOWN);
System.out.println(result);
Use a special method for divide. Normal / cannot be used. See oracle.docs for details.
How to handle 0 of BigDecimal in conditional branching etc. I'm addicted to this if I'm not used to it. Of course I was addicted to it because I am also a beginner.
ZeroCompare.java
if(Value to compare.compareTo(BigDecimal.ZERO) == 0) {
//・ ・ ・
}
It looks like. In other words, there is no difference compared to 0, so 0! It will be a comparison. By the way, if it is 0 or more or 0 or less
Compare.java
ZeroCompare.java
if(Value to compare.compareTo(BigDecimal.ZERO) > 0) {
//・ ・ ・
}
To do.
BigDecimal is used differently than primitive types. It is necessary to be aware of that. Especially for java beginners like me, it's hard to accidentally use floats and doubles when you have to make accurate calculations in actual development work, or to use BigDecimal incorrectly and cause compilation errors. Don't do it. You may not usually care about it, but in the case of work, I would like to listen carefully to the purpose of the calculation logic to be developed and the required accuracy before developing it. [Reflection]
Recommended Posts