NSDecimalNumber 四舍五入到指定位数

发表于   |     |   阅读次数

项目里经常有这么一种需求，把小数四舍五入到指定位数，查阅文档，我们发现系统提供了NSDecimalNumber类方便我们实现需求。
NSDecimalNumber是系统提供的一个专门处理十进位制数的类：

1
NSDecimalNumber, an immutable subclass of NSNumber, provides an object-oriented wrapper for doing base-10 arithmetic. An instance can represent any number that can be expressed as mantissa x 10^exponent where mantissa is a decimal integer up to 38 digits long, and exponent is an integer from –128 through 127.

NSDecimalNumber提供了处理舍入问题的方法：

1
2
3
Rounds the receiver off in the way specified by behavior and returns the result, a newly createdNSDecimalNumber object.

- (NSDecimalNumber *)decimalNumberByRoundingAccordingToBehavior:(nullable id <NSDecimalNumberBehaviors>)behavior;

该方法需要调用者提供一个遵循NSDecimalNumerBehaviors协议的执行者来处理舍入问题。这里我们无需自定义一个类，只需使用系统提供的NSDecimalNumberHandler类来处理。该类已遵循NSDecimalNumerBehaviors协议，且提供了一些方法方便我们实现：

1
NSDecimalNumberHandler is a class that adopts the NSDecimalNumberBehaviors protocol. This class allows you to set the way an NSDecimalNumber object rounds off and handles errors, without having to create a custom class.

NSDecimalNumberHandler提供的舍入处理方法：

1
2
3
Returns an NSDecimalNumberHandler object with customized behavior.

+ (instancetype)decimalNumberHandlerWithRoundingMode:(NSRoundingMode)roundingMode scale:(short)scale raiseOnExactness:(BOOL)exact raiseOnOverflow:(BOOL)overflow raiseOnUnderflow:(BOOL)underflow raiseOnDivideByZero:(BOOL)divideByZero;

该方法需要我们设置六个参数，第一个参数为舍入模式(roundingMode)，第二个参数为保留位数(scale，以小数点为中心，正数为小数位，负数为整数位)，第三个参数为数值精度异常捕获(raiseOnExactness)，第四个参数为数值上溢异常捕获(raiseOnOverflow)，第五个参数为数值下溢异常捕获(raiseOnUnderflow)，最后一个参数为数值除数为零异常捕获(raiseOnDivideByZero)：

1
2
3
4
5
6
7
8
9
10
11
roundingMode:The rounding mode to use. There are four possible values: NSRoundUp,NSRoundDown, NSRoundPlain, and NSRoundBankers.

scale:The number of digits a rounded value should have after its decimal point.

raiseOnExactness:If YES, in the event of an exactness error the handler will raise an exception, otherwise it will ignore the error and return control to the calling method.

raiseOnOverflow:If YES, in the event of an overflow error the handler will raise an exception, otherwise it will ignore the error and return control to the calling method.

raiseOnUnderflow:If YES, in the event of an underflow error the handler will raise an exception, otherwise it will ignore the error and return control to the calling method.

raiseOnDivideByZero:If YES, in the event of a divide by zero error the handler will raise an exception, otherwise it will ignore the error and return control to the calling method.

参数可根据实际需求设定，需要注意的是舍入模式(roundingMode)分别有四种，在这里我们使用NSRoundPlain模式：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
// Rounding policies :
// Original
// value    1.2  1.21  1.25  1.35  1.27
// Plain    1.2  1.2   1.3   1.4   1.3
// Down     1.2  1.2   1.2   1.3   1.2
// Up       1.2  1.3   1.3   1.4   1.3
// Bankers  1.2  1.2   1.2   1.4   1.3

typedef NS_ENUM(NSUInteger, NSRoundingMode) {
    NSRoundPlain,   // Round up on a tie
    NSRoundDown,    // Always down == truncate
    NSRoundUp,      // Always up
    NSRoundBankers  // on a tie round so last digit is even
};

基本要素齐备后，可以进行方法接口设计。在这里我们采用NSDecimalNumber分类方法的形式实现。为NSDecimalNumber提供舍入方法，根据舍入模式舍入到指定位数：

1
2
3
4
5
- (NSDecimalNumber *)roundToScale:(short)scale mode:(NSRoundingMode)roundingMode
{
    NSDecimalNumberHandler *handler = [NSDecimalNumberHandler decimalNumberHandlerWithRoundingMode:roundingMode scale:scale raiseOnExactness:NO raiseOnOverflow:YES raiseOnUnderflow:YES raiseOnDivideByZero:YES];
    return [self decimalNumberByRoundingAccordingToBehavior:handler];
}

四舍五入到指定位数：

1
2
3
4
- (NSDecimalNumber *)roundToScale:(short)scale
{
    return [self roundToScale:scale mode:NSRoundPlain];
}

为float和double类型数值添加创建方法。根据float数值、舍入位数和舍入模式, 创建decimalNumber：

1
2
3
4
+ (NSDecimalNumber *)decimalNumberWithFloat:(float)value roundingScale:(short)scale roundingMode:(NSRoundingMode)mode
{
    return [[[NSDecimalNumber alloc] initWithFloat:value] roundToScale:scale mode:mode];
}

根据float数值和四舍五入位数, 创建decimalNumber：

1
2
3
4
+ (NSDecimalNumber *)decimalNumberWithFloat:(float)value roundingScale:(short)scale
{
    return [[[NSDecimalNumber alloc] initWithFloat:value] roundToScale:scale];
}

根据double数值、舍入位数和舍入模式, 创建decimalNumber：

1
2
3
4
+ (NSDecimalNumber *)decimalNumberWithDouble:(double)value roundingScale:(short)scale roundingMode:(NSRoundingMode)mode
{
    return [[[NSDecimalNumber alloc] initWithDouble:value] roundToScale:scale mode:mode];
}

根据double数值和四舍五入位数, 创建decimalNumber：

1
2
3
4
+ (NSDecimalNumber *)decimalNumberWithDouble:(double)value roundingScale:(short)scale
{
    return [[[NSDecimalNumber alloc] initWithDouble:value] roundToScale:scale];
}

至此，我们在需要四舍五入数值时调用即可：

1
2
// 四舍五入至小数两位
NSDecimalNumber *decimalNumber = [NSDecimalNumber decimalNumberWithFloat:2.567f roundingScale:2];

以上方法适用于一般四舍五入需求，然而有时候我们可能面对这样的需求，任意数值四舍五入至指定小数位后且保留并显示至指定小数位，不足的补0显示(例，1.445->1.45; 1.4->1.40)。而以上方法默认输出数值是舍去尾数0的。对此，我们有两种方法实现，第一种是四舍五入后返回decimalNumber的floatValue或doubleValue后再处理：

1
2
3
4
5
6
// 四舍五入并显示至小数后两位, 不足0补齐
NSDecimalNumber *decimalNumber = [NSDecimalNumber decimalNumberWithFloat:2.5f roundingScale:2];
NSLog(@"%.2f", [decimalNumber floatValue]);

NSDecimalNumber *decimalNumber = [NSDecimalNumber decimalNumberWithDouble:2.5 roundingScale:2];
NSLog(@"%.2f", [decimalNumber doubleValue]);

该方法虽然简单，但是会破坏数值精度。若想不破话数值精度，则可以通过NSNumberFomatter实现：

1
2
3
4
5
6
// 四舍五入并显示至小数后两位, 不足0补齐
NSDecimalNumber *decimalNumber = [NSDecimalNumber decimalNumberWithFloat:value roundingScale:2];
NSNumberFormatter *numberFormatter = [NSNumberFormatter new];
[numberFormatter setMaximumFractionDigits:2];
[numberFormatter setMinimumFractionDigits:2];
NSLog(@"%@", [numberFormatter stringFromNumber:decimalNumber]);

同时，我们可以通过NSString分类方法进行扩展，方便实现：
根据number和保留显示小数位数，获取对应的字符串：

1
2
3
4
5
6
7
+ (NSString *)stringFromNumber:(NSNumber *)number fractionDigits:(NSUInteger)fractionDigits
{
    NSNumberFormatter *numberFormatter = [NSNumberFormatter new];
    [numberFormatter setMaximumFractionDigits:fractionDigits];
    [numberFormatter setMinimumFractionDigits:fractionDigits];
    return [numberFormatter stringFromNumber:number];
}

根据float数值、舍入位数、舍入模式和保留显示小数位数，获取对应的字符串：

1
2
3
4
5
+ (NSString *)stringFromFloat:(float)value roundingScale:(short)scale roundingMode:(NSRoundingMode)mode fractionDigits:(NSUInteger)fractionDigits
{
    NSDecimalNumber *decimalNumber = [NSDecimalNumber decimalNumberWithFloat:value roundingScale:scale roundingMode:mode];
    return [NSString stringFromNumber:decimalNumber fractionDigits:fractionDigits];
}

根据float数值、舍入位数、舍入模式和是否补齐显示小数位数(不足补0，补齐位数与四舍五入位数一致)，获取对应的字符串：

1
2
3
4
5
6
7
8
+ (NSString *)stringFromFloat:(float)value roundingScale:(short)scale roundingMode:(NSRoundingMode)mode fractionDigitsPadded:(BOOL)isPadded
{
    NSDecimalNumber *decimalNumber = [NSDecimalNumber decimalNumberWithFloat:value roundingScale:scale roundingMode:mode];

    if (!isPadded) return [NSString stringWithFormat:@"%@", decimalNumber];

    return [NSString stringFromNumber:decimalNumber fractionDigits:scale];
}

根据float数值、四舍五入位数和是否补齐显示小数位数(不足补0，补齐位数与四舍五入位数一致)，获取对应的字符串：

1
2
3
4
+ (NSString *)stringFromFloat:(float)value roundingScale:(short)scale fractionDigitsPadded:(BOOL)isPadded
{
    return [NSString stringFromFloat:value roundingScale:scale roundingMode:NSRoundPlain fractionDigitsPadded:isPadded];
}

根据float数值和四舍五入位数，获取对应的字符串：

1
2
3
4
+ (NSString *)stringFromFloat:(float)value roundingScale:(short)scale
{
    return [NSString stringFromFloat:value roundingScale:scale roundingMode:NSRoundPlain fractionDigitsPadded:NO];
}

double数值的实现与float类似。在需要的位置如下调用即可：

1
2
3
4
5
6
// 四舍五入并显示至小数后两位, 不足0补齐
NSDecimalNumber *decimalNumber = [NSDecimalNumber decimalNumberWithFloat:2.4 roundingScale:2];
NSLog(@"%@", [NSString stringFromNumber:decimalNumber fractionDigits:2]);

// 四舍五入并显示至小数后两位, 不足0补齐, 补齐位数与四舍五入位数一致
NSLog(@"%@", [NSString stringFromFloat:2.4 roundingScale:2 fractionDigitsPadded:YES]);

附，示例代码：https://github.com/ColinHwang/iOS-RoundNumber

参考资料：

[1]iOS-Denzel.关于OC中的小数精确计算—NSDecimalNumber[EB/OL].http://www.cnblogs.com/denz/p/5330771.html ,2016-3-28
[2]Apple Inc.NSDecimalNumber Classs Reference[EB/OL].https://developer.apple.com/library/mac/documentation/Cocoa/Reference/Foundation/Classes/NSDecimalNumber_Class/index.html ,2016-6-22.
[3]Apple Inc.NSNumberFormatter Classs Reference[EB/OL].https://developer.apple.com/library/mac/documentation/Cocoa/Reference/Foundation/Classes/NSNumberFormatter_Class/index.html ,2016-6-22.