Documentation ¶
Overview ¶
Package big extends the capabilities of standard "math/big" package by adding custom global precision to Float and Rat; and global rounding mode, and bits precision to Float.
Index ¶
- Variables
- type Float
- func AddFloat(f ...interface{}) *Float
- func CreateFloat(v float64) Float
- func MulFloat(f, g interface{}) *Float
- func MustParseFloat(s string) (f *Float)
- func NewFloat(v interface{}) *Float
- func ParseFloat(s string) (f *Float, err error)
- func QuoFloat(f, g interface{}) *Float
- func SubFloat(f, g *Float) *Float
- func (f *Float) Add(g interface{}) *Float
- func (f *Float) Clone() *Float
- func (f *Float) Int64() int64
- func (f *Float) IsEqual(g interface{}) bool
- func (f *Float) IsGreater(g interface{}) bool
- func (f *Float) IsGreaterOrEqual(g interface{}) bool
- func (f *Float) IsLess(g interface{}) bool
- func (f *Float) IsLessOrEqual(g interface{}) bool
- func (f *Float) IsZero() bool
- func (f *Float) MarshalJSON() ([]byte, error)
- func (f *Float) Mul(g interface{}) *Float
- func (f *Float) ParseFloat(s string) (err error)
- func (f *Float) Quo(g interface{}) *Float
- func (f *Float) String() string
- func (f *Float) Sub(g interface{}) *Float
- func (f *Float) UnmarshalJSON(in []byte) (err error)
- type Int
- func (i *Int) Add(v interface{}) *Int
- func (i *Int) IsGreater(v interface{}) bool
- func (i *Int) IsLess(v interface{}) bool
- func (i *Int) IsZero() bool
- func (i *Int) MarshalJSON() ([]byte, error)
- func (i *Int) Scan(src interface{}) error
- func (i *Int) UnmarshalJSON(in []byte) (err error)
- func (i *Int) Value() (driver.Value, error)
- type Rat
- func (r *Rat) Abs() *Rat
- func (r *Rat) Add(g interface{}) *Rat
- func (r *Rat) Humanize(thousandSep, decimalSep string) string
- func (r *Rat) Int64() int64
- func (r *Rat) IsEqual(g interface{}) bool
- func (r *Rat) IsGreater(g interface{}) bool
- func (r *Rat) IsGreaterOrEqual(g interface{}) bool
- func (r *Rat) IsGreaterThanZero() bool
- func (r *Rat) IsLess(g interface{}) bool
- func (r *Rat) IsLessOrEqual(g interface{}) bool
- func (r *Rat) IsLessThanZero() bool
- func (r *Rat) IsZero() bool
- func (r *Rat) MarshalJSON() ([]byte, error)
- func (r *Rat) Mul(g interface{}) *Rat
- func (r *Rat) Quo(g interface{}) *Rat
- func (r *Rat) RoundNearestFraction() *Rat
- func (r *Rat) RoundToNearestAway(prec int) *Rat
- func (r *Rat) RoundToZero(prec int) *Rat
- func (r *Rat) Scan(v interface{}) error
- func (r *Rat) String() string
- func (r *Rat) Sub(g interface{}) *Rat
- func (r *Rat) UnmarshalJSON(in []byte) (err error)
- func (r *Rat) Value() (driver.Value, error)
Examples ¶
- AddRat
- MulRat
- NewRat
- QuoRat
- Rat.Abs
- Rat.Add
- Rat.Humanize
- Rat.Int64
- Rat.IsEqual
- Rat.IsGreater
- Rat.IsGreaterOrEqual
- Rat.IsGreaterThanZero
- Rat.IsLess
- Rat.IsLessOrEqual
- Rat.IsLessThanZero
- Rat.IsZero
- Rat.MarshalJSON
- Rat.MarshalJSON (WithStruct)
- Rat.Mul
- Rat.Quo
- Rat.RoundNearestFraction
- Rat.RoundToNearestAway
- Rat.RoundToZero
- Rat.Scan
- Rat.String
- Rat.Sub
- Rat.UnmarshalJSON (WithStruct)
- Rat.Value
- SubRat
Constants ¶
This section is empty.
Variables ¶
var DefaultBitPrecision uint = 128
DefaultBitPrecision define the maximum number of mantissa bits available to represent the value.
In standard library this value is 24 for float32 or 53 for float64.
One should change this value before using the new extended Float in the program.
var DefaultDigitPrecision = 8
DefaultDigitPrecision define the default number of digits after decimal point which affect the return of String() and MarshalJSON() methods.
A zero value of digit precision mean is it will use the default output of 'f' format.
One should change this value before using the new extended Float or Rat in the program.
var DefaultRoundingMode = big.ToNearestAway
DefaultRoundingMode define the default rounding mode for all instance of Float.
One should change this value before using the new extended Float in the program.
var MarshalJSONAsString bool = true
MarshalJSONAsString define the default return behaviour of MarshalJSON(). If its true (the default) the returned JSON format will encapsulated in double quote, as string instead of as numeric.
Functions ¶
This section is empty.
Types ¶
type Float ¶
Float extend the standard big.Float by setting each instance precision to DefaultBitPrecision, rounding mode to DefaultRoundingMode, and using DefaultDigitPrecision value after decimal point when converted to string.
func AddFloat ¶
func AddFloat(f ...interface{}) *Float
AddFloat return the rounded sum `f[0]+f[1]+...`. It will return nil if the first parameter is not convertable to Float.
func CreateFloat ¶
Create Float with default bit precision and rounding mode.
func MulFloat ¶
func MulFloat(f, g interface{}) *Float
MulFloat return the result of multiplication `f*g`. It will return nil if `f` or `g` is not convertible to Float.
func MustParseFloat ¶
MustParseFloat convert the string `s` into Float or panic.
func NewFloat ¶
func NewFloat(v interface{}) *Float
NewFloat create and initialize new Float with default bit precision, and rounding mode.
func ParseFloat ¶
ParseFloat the string s into Float value.
func QuoFloat ¶
func QuoFloat(f, g interface{}) *Float
QuoFloat return the quotient of `f/g` as new Float.
func (*Float) IsGreaterOrEqual ¶
IsGreaterOrEqual will return true if `f >= g`.
func (*Float) IsLessOrEqual ¶
IsLessOrEqual will return true if `f <= g`.
func (*Float) MarshalJSON ¶
MarshalJSON implement the json.Marshaler interface and return the output of String method.
func (*Float) Mul ¶
Mul sets f to product of `f * g` and return the result as f. If g is not convertible to Float it will return nil.
func (*Float) ParseFloat ¶
Parse the string into Float value.
func (*Float) Quo ¶
Quo sets f to quotient of `f/g` and return the result as f. If g is not convertible to Float it will return nil.
func (*Float) String ¶
String format the Float value into string with maximum mantissa is set by digit precision option.
Unlike standard String method, this method will trim trailing zero digit or decimal point at the end of mantissa.
func (*Float) UnmarshalJSON ¶
UnmarshalJSON convert the JSON byte value into Float.
type Int ¶ added in v0.26.0
Int extends the standard big.Int package.
func NewInt ¶ added in v0.26.0
func NewInt(v interface{}) (i *Int)
NewInt create and initialize new Int value from v or nil if v is invalid type that cannot be converted to Int.
func (*Int) MarshalJSON ¶ added in v0.26.0
MarshalJSON implement the json.Marshaler interface and return the output of String method.
If the global variable MarshalJSONAsString is true, the Int value will be encoded as string.
func (*Int) UnmarshalJSON ¶ added in v0.26.0
UnmarshalJSON convert the JSON byte value into Int.
type Rat ¶
Rat extend the standard big.Rat using rounding mode ToZero and without panic.
func AddRat ¶
func AddRat(f ...interface{}) *Rat
AddRat return the sum of `f[0]+f[1]+...`. It will return nil if the first parameter is not convertable to Rat.
Example ¶
fmt.Println(AddRat()) fmt.Println(AddRat(nil)) fmt.Println(AddRat("a")) fmt.Println(AddRat(0, 0.0001)) fmt.Println(AddRat("1.007", "a", "2.003")) // Invalid parameter "a" is ignored.
Output: 0 0 0 0.0001 3.01
func MulRat ¶
func MulRat(f ...interface{}) *Rat
MulRat return the result of multiplication `f[0]*f[1]*...`. It will return nil if the first parameter is not convertable to Rat.
Example ¶
fmt.Println(MulRat()) fmt.Println(MulRat(nil)) fmt.Println(MulRat("a")) fmt.Println(MulRat(0, 1)) fmt.Println(MulRat(6, "a", "0.3")) // Invalid parameter "a" is ignored.
Output: 0 0 0 0 1.8
func NewRat ¶
func NewRat(v interface{}) (r *Rat)
NewRat create and initialize new Rat value from v. It will return nil if v is not convertable to Rat.
Empty string or empty []byte still considered as valid, and it will return it as zero.
Example ¶
var ( stdNilInt *big.Int stdNilRat *big.Rat stdRat big.Rat nilRat *Rat ) inputs := []interface{}{ nil, []byte{}, "", []byte("a"), "0", "0.0000_0001", []byte("14687233442.06916608"), "14_687_233_442.069_166_08", nilRat, NewRat("14687233442.06916608"), *NewRat("14687233442.06916608"), stdNilRat, stdRat, big.NewRat(14687233442, 100_000_000), *big.NewRat(14687233442, 100_000_000), uint16(math.MaxUint16), uint32(math.MaxUint32), uint64(math.MaxUint64), stdNilInt, big.NewInt(100_000_000), } for _, v := range inputs { fmt.Println(NewRat(v)) }
Output: 0 0 0 0 0 0.00000001 14687233442.06916608 14687233442.06916608 0 14687233442.06916608 14687233442.06916608 0 0 146.87233442 146.87233442 65535 4294967295 18446744073709551615 0 100000000
func QuoRat ¶
func QuoRat(f ...interface{}) *Rat
QuoRat return the quotient of `f[0]/f[1]/...` as new Rat. It will return nil if the first parameter is not convertable to Rat. If the second or rest of parameters can not be converted to Rat or zero it will return nil instead of panic.
Example ¶
fmt.Println(QuoRat()) fmt.Println(QuoRat(nil)) fmt.Println(QuoRat("a")) fmt.Println(QuoRat("0")) fmt.Println(QuoRat(2, 0, 2)) fmt.Println(QuoRat(6, "a", "0.3")) fmt.Println(QuoRat(0, 1)) fmt.Println(QuoRat(4651, 272)) fmt.Println(QuoRat(int64(1815507979407), NewRat(100000000))) fmt.Println(QuoRat("25494300", "25394000000"))
Output: 0 0 0 0 0 0 0 17.0992647 18155.07979407 0.00100395
func SubRat ¶
func SubRat(f ...interface{}) *Rat
SubRat return the result of subtraction `f[0]-f[1]-...` as new Rat. It will return nil if the first parameter is not convertable to Rat.
Example ¶
fmt.Println(SubRat()) fmt.Println(SubRat(nil)) fmt.Println(SubRat("a")) fmt.Println(SubRat(0, 1)) fmt.Println(SubRat(6, "a", "0.3"))
Output: 0 0 0 -1 5.7
func (*Rat) Abs ¶ added in v0.16.0
Abs sets r to |r| (the absolute value of r) and return it.
Example ¶
fmt.Println(NewRat(nil).Abs()) fmt.Println(NewRat("-1").Abs()) fmt.Println(NewRat("-0.00001").Abs()) fmt.Println(NewRat("1").Abs())
Output: 0 1 0.00001 1
func (*Rat) Add ¶
Add sets r to `r+g` and return the r as the result. If g is not convertable to Rat it will equal to r+0.
Example ¶
fmt.Println(NewRat(nil).Add(nil)) fmt.Println(NewRat(1).Add(nil)) fmt.Println(NewRat(1).Add(1))
Output: 0 1 2
func (*Rat) Humanize ¶ added in v0.23.0
Humanize format the r into string with custom thousand and decimal separator.
Example ¶
var ( thousandSep = "." decimalSep = "," ) fmt.Printf("%s\n", NewRat(nil).Humanize(thousandSep, decimalSep)) fmt.Printf("%s\n", NewRat("0").Humanize(thousandSep, decimalSep)) fmt.Printf("%s\n", NewRat("0.1234").Humanize(thousandSep, decimalSep)) fmt.Printf("%s\n", NewRat("100").Humanize(thousandSep, decimalSep)) fmt.Printf("%s\n", NewRat("100.1234").Humanize(thousandSep, decimalSep)) fmt.Printf("%s\n", NewRat("1000").Humanize(thousandSep, decimalSep)) fmt.Printf("%s\n", NewRat("1000.2").Humanize(thousandSep, decimalSep)) fmt.Printf("%s\n", NewRat("10000.23").Humanize(thousandSep, decimalSep)) fmt.Printf("%s\n", NewRat("100000.234").Humanize(thousandSep, decimalSep))
Output: 0 0 0,1234 100 100,1234 1.000 1.000,2 10.000,23 100.000,234
func (*Rat) Int64 ¶
Int64 return the integer resulting from truncating r towards zero. It will return math.MaxInt64, if the value is larger than MaxInt64. It will return math.MinInt64, if the value is lower than MinInt64.
Example ¶
fmt.Printf("MaxInt64: %d\n", NewRat("9223372036854775807").Int64()) fmt.Printf("MaxInt64+1: %d\n", NewRat("9223372036854775808").Int64()) fmt.Printf("MinInt64: %d\n", NewRat("-9223372036854775808").Int64()) fmt.Printf("MinInt64-1: %d\n", NewRat("-9223372036854775809").Int64()) fmt.Println(NewRat(nil).Int64()) fmt.Println(NewRat("0.000_000_001").Int64()) fmt.Println(NewRat("0.5").Int64()) fmt.Println(NewRat("0.6").Int64()) fmt.Println(NewRat("4011144900.02438879").Mul(100000000).Int64()) fmt.Println(QuoRat("128_900", "0.000_0322").Int64()) fmt.Println(QuoRat(128900, 3220).Mul(100000000).Int64()) fmt.Println(QuoRat(25494300, QuoRat(25394000000, 100000000)).Int64())
Output: MaxInt64: 9223372036854775807 MaxInt64+1: 9223372036854775807 MinInt64: -9223372036854775808 MinInt64-1: -9223372036854775808 0 0 0 0 401114490002438879 4003105590 4003105590 100394
func (*Rat) IsEqual ¶
IsEqual will return true if `r == g`, including when r and g are both nil.
Unlike the standard Cmp(), if the first call to Cmp is not 0, it will try to compare the string values of r and g, truncated by DefaultDigitPrecision.
Example ¶
f := NewRat(1) fmt.Println(NewRat(nil).IsEqual(0)) fmt.Println(f.IsEqual("a")) fmt.Println(f.IsEqual("1")) fmt.Println(f.IsEqual(1.1)) fmt.Println(f.IsEqual(byte(1))) fmt.Println(f.IsEqual(int(1))) fmt.Println(f.IsEqual(int32(1))) fmt.Println(f.IsEqual(float32(1))) fmt.Println(f.IsEqual(NewRat(1))) // Equal due to String() truncation to DefaultDigitPrecision (8) digits. fmt.Println(NewRat("0.1234567890123").IsEqual("0.12345678"))
Output: false false true false true true true true true true
func (*Rat) IsGreater ¶
IsGreater will return true if `r > g`. If g is not convertable to Rat it will return false.
Example ¶
r := NewRat("0.000_000_5") fmt.Println(NewRat(nil).IsGreater(0)) fmt.Println(r.IsGreater(nil)) fmt.Println(r.IsGreater("0.000_000_5")) fmt.Println(r.IsGreater("0.000_000_49999"))
Output: false false false true
func (*Rat) IsGreaterOrEqual ¶
IsGreaterOrEqual will return true if `r >= g`. If g is not convertable to Rat it will return false.
Example ¶
r := NewRat("0.000_000_5") fmt.Println(NewRat(nil).IsGreaterOrEqual(0)) fmt.Println(r.IsGreaterOrEqual(nil)) fmt.Println(r.IsGreaterOrEqual("0.000_000_500_000_000_001")) fmt.Println(r.IsGreaterOrEqual("0.000_000_5")) fmt.Println(r.IsGreaterOrEqual("0.000_000_49999"))
Output: false false false true true
func (*Rat) IsGreaterThanZero ¶
IsGreaterThanZero will return true if `r > 0`.
Example ¶
fmt.Println(NewRat(nil).IsGreaterThanZero()) fmt.Println(NewRat(0).IsGreaterThanZero()) fmt.Println(NewRat("-0.000_000_000_000_000_001").IsGreaterThanZero()) fmt.Println(NewRat("0.000_000_000_000_000_001").IsGreaterThanZero())
Output: false false false true
func (*Rat) IsLess ¶
IsLess will return true if `r < g`. If r is nill or g is not convertable to Rat it will return false.
Example ¶
r := NewRat("0.000_000_5") fmt.Println(NewRat(nil).IsLess(0)) fmt.Println(r.IsLess(nil)) fmt.Println(r.IsLess("0.000_000_5")) fmt.Println(r.IsLess("0.000_000_49")) fmt.Println(r.IsLess("0.000_000_500_000_000_001"))
Output: false false false false true
func (*Rat) IsLessOrEqual ¶
IsLessOrEqual will return true if `r <= g`. It r is nil or g is not convertable to Rat it will return false.
Example ¶
r := NewRat("0.000_000_5") fmt.Println(NewRat(nil).IsLessOrEqual(r)) fmt.Println(r.IsLessOrEqual(nil)) fmt.Println(r.IsLessOrEqual("0.000_000_5")) fmt.Println(r.IsLessOrEqual("0.000_000_49")) fmt.Println(r.IsLessOrEqual("0.000_000_500_000_000_001"))
Output: false false true false true
func (*Rat) IsLessThanZero ¶
IsLessThanZero return true if `r < 0`.
Example ¶
fmt.Println(NewRat(nil).IsLessThanZero()) fmt.Println(NewRat(byte(0)).IsLessThanZero()) fmt.Println(NewRat("-0.000_000_000_000_000_001").IsLessThanZero()) fmt.Println(NewRat("0.000_000_000_000_000_001").IsLessThanZero())
Output: false false true false
func (*Rat) IsZero ¶
IsZero will return true if `r == 0`.
Example ¶
fmt.Println(NewRat(nil).IsZero()) fmt.Println(NewRat(byte(0)).IsZero()) fmt.Println(NewRat(byte(-0)).IsZero()) fmt.Println(NewRat("-0.000_000_000_000_000_001").IsZero()) fmt.Println(NewRat("0.000_000_000_000_000_001").IsZero())
Output: false true true false false
func (*Rat) MarshalJSON ¶
MarshalJSON implement the json.Marshaler interface. It will return the same result as String().
Example ¶
inputs := []string{ "", "a", "0.0000_0000", "0.1", "0.0000_0001", "0.0000_0000_1", // Truncated by DefaultDigitPrecision. "1234567890.0", "64.23738872403", // Truncated by DefaultDigitPrecision. "0.1234567890", "142660378.65368736", "9193394308.85771370", "14687233442.06916608", } MarshalJSONAsString = true for _, in := range inputs { out, _ := NewRat(in).MarshalJSON() fmt.Printf("%s\n", out) } // Setting this to false will make the JSON output become number. MarshalJSONAsString = false for _, in := range inputs { out, _ := NewRat(in).MarshalJSON() fmt.Printf("%s\n", out) }
Output: "0" "0" "0" "0.1" "0.00000001" "0" "1234567890" "64.23738872" "0.12345678" "142660378.65368736" "9193394308.8577137" "14687233442.06916608" 0 0 0 0.1 0.00000001 0 1234567890 64.23738872 0.12345678 142660378.65368736 9193394308.8577137 14687233442.06916608
Example (WithStruct) ¶
type T struct { V *Rat } inputs := []T{ {V: nil}, {V: NewRat(0)}, {V: NewRat("0.1234567890")}, } MarshalJSONAsString = true for _, in := range inputs { out, _ := json.Marshal(&in) fmt.Printf("%s\n", out) } MarshalJSONAsString = false for _, in := range inputs { out, _ := json.Marshal(&in) fmt.Printf("%s\n", out) }
Output: {"V":null} {"V":"0"} {"V":"0.12345678"} {"V":null} {"V":0} {"V":0.12345678}
func (*Rat) Mul ¶
Mul sets r to product of `r * g` and return the result as r. If g is not convertible to Rat it will return nil.
Example ¶
const ( defValue = "14687233442.06916608" ) fmt.Println(NewRat(defValue).Mul("a")) fmt.Println(NewRat(defValue).Mul("0")) fmt.Println(NewRat(defValue).Mul(defValue)) fmt.Println(NewRat("1.06916608").Mul("1.06916608"))
Output: 0 0 215714826181834884090.46087866 1.1431161
func (*Rat) Quo ¶
Quo sets r to quotient of `r/g` and return the result as r. If r is nil or g is not convertible to Rat or zero it will return nil.
Example ¶
const ( defValue = "14687233442.06916608" ) fmt.Println(NewRat(nil).Quo(1)) fmt.Println(NewRat(defValue).Quo(nil)) fmt.Println(NewRat(defValue).Quo("a")) fmt.Println(NewRat(defValue).Quo("100_000_000"))
Output: 0 0 0 146.87233442
func (*Rat) RoundNearestFraction ¶ added in v0.23.0
RoundNearestFraction round the fraction to the nearest non-zero value.
The RoundNearestFraction does not require precision parameter, like in other rounds function, but it will figure it out based on the last non-zero value from fraction.
See example for more information.
Example ¶
fmt.Printf("nil: %s\n", NewRat(nil).RoundNearestFraction()) fmt.Printf("0.000000001: %s\n", NewRat("0").RoundNearestFraction()) // Affected by DefaultDigitPrecision (8) fmt.Printf("0.00545: %s\n", NewRat("0.00545").RoundNearestFraction()) fmt.Printf("0.00555: %s\n", NewRat("0.00555").RoundNearestFraction()) fmt.Printf("0.0545: %s\n", NewRat("0.0545").RoundNearestFraction()) fmt.Printf("0.0555: %s\n", NewRat("0.0555").RoundNearestFraction()) fmt.Printf("0.545: %s\n", NewRat("0.545").RoundNearestFraction()) fmt.Printf("0.555: %s\n", NewRat("0.555").RoundNearestFraction()) fmt.Printf("0.5: %s\n", NewRat("0.5").RoundNearestFraction()) fmt.Printf("-0.5: %s\n", NewRat("-0.5").RoundNearestFraction()) fmt.Printf("-0.555: %s\n", NewRat("-0.555").RoundNearestFraction()) fmt.Printf("-0.545: %s\n", NewRat("-0.545").RoundNearestFraction())
Output: nil: 0 0.000000001: 0 0.00545: 0.005 0.00555: 0.006 0.0545: 0.05 0.0555: 0.06 0.545: 0.5 0.555: 0.6 0.5: 0.5 -0.5: -0.5 -0.555: -0.6 -0.545: -0.5
func (*Rat) RoundToNearestAway ¶ added in v0.20.1
RoundToNearestAway round r to n digit precision using nearest away mode, where mantissa is accumulated by the last digit after precision. For example, using 2 digit precision, 0.555 would become 0.56.
Example ¶
fmt.Printf("nil: %s\n", NewRat(nil).RoundToNearestAway(2)) fmt.Printf("0.0054: %s\n", NewRat("0.0054").RoundToNearestAway(2)) fmt.Printf("0.0054: %s\n", NewRat("0.0054").RoundToNearestAway(1)) fmt.Printf("0.5455: %s\n", NewRat("0.5455").RoundToNearestAway(2)) fmt.Printf("0.5555: %s\n", NewRat("0.5555").RoundToNearestAway(2)) fmt.Printf("0.5566: %s\n", NewRat("0.5567").RoundToNearestAway(2)) fmt.Printf("0.5566: %s\n", NewRat("0.5566").RoundToNearestAway(0)) fmt.Printf("0.02514135: %s\n", NewRat("0.02514135").RoundToNearestAway(6)) fmt.Printf("0.02514145: %s\n", NewRat("0.02514145").RoundToNearestAway(6)) fmt.Printf("0.02514155: %s\n", NewRat("0.02514155").RoundToNearestAway(6)) fmt.Printf("0.02514165: %s\n", NewRat("0.02514165").RoundToNearestAway(6)) fmt.Printf("0.5: %s\n", NewRat("0.5").RoundToNearestAway(0)) fmt.Printf("-0.5: %s\n", NewRat("-0.5").RoundToNearestAway(0))
Output: nil: 0 0.0054: 0.01 0.0054: 0 0.5455: 0.55 0.5555: 0.56 0.5566: 0.56 0.5566: 1 0.02514135: 0.025141 0.02514145: 0.025141 0.02514155: 0.025142 0.02514165: 0.025142 0.5: 1 -0.5: -1
func (*Rat) RoundToZero ¶ added in v0.20.1
RoundToZero round r to n digit precision using to zero mode. For example, using 2 digit precision, 0.555 would become 0.55.
Example ¶
fmt.Println(NewRat(nil).RoundToZero(2)) fmt.Println(NewRat("0.5455").RoundToZero(2)) fmt.Println(NewRat("0.5555").RoundToZero(2)) fmt.Println(NewRat("0.5566").RoundToZero(2)) fmt.Println(NewRat("0.5566").RoundToZero(0)) // In Go <= 1.18, this will print "-0", but on Go tip "0". // So to make test success on all versions, we multiple it to 1. fmt.Println(NewRat("-0.5").RoundToZero(0).Mul(1))
Output: 0 0.54 0.55 0.55 0 0
func (*Rat) Scan ¶
Scan implement the database's sql.Scan interface.
Example ¶
var ( r = &Rat{} err error ) inputs := []interface{}{ 1234, nil, "0.0001", float64(0.0001), } for _, in := range inputs { err = r.Scan(in) if err != nil { fmt.Printf("error: %s\n", err) } fmt.Println(r) }
Output: 1234 error: Rat.Scan: unknown type <nil> 1234 0.0001 0.0001
func (*Rat) String ¶
String format the Rat value into string with maximum mantissa is set by digit precision option with rounding mode set to zero.
Unlike standard String method, this method will trim trailing zero digit or decimal point at the end of mantissa.
Example ¶
inputs := []interface{}{ nil, "12345", "0.00000000", float64(0.00000000), "0.1", float64(0.1), "0.0000001", float64(0.0000001), "0.000000001", // Truncated due to rounding. "64.23738872403", float64(64.23738872403), }
Output: 0 12345 0 0 0.1 0.1 0.0000001 0.0000001 0 64.23738872 64.23738872
func (*Rat) Sub ¶
Sub sets r to rounded difference `r-g` and return r. If g is not convertable to Rat, it will return as r-0.
Example ¶
fmt.Println(NewRat(nil).Sub(1)) fmt.Println(NewRat(1).Sub(nil)) fmt.Println(NewRat(1).Sub(1))
Output: 0 1 0
func (*Rat) UnmarshalJSON ¶
UnmarshalJSON convert the JSON byte value into Rat.
Example (WithStruct) ¶
type T struct { V *Rat `json:"V"` W *Rat `json:"W,omitempty"` } inputs := []string{ `{"V":"ab"}`, `{}`, `{"V":}`, `{"V":0,"W":0}`, `{"V":"1"}`, `{"V":0.123456789}`, `{"V":"0.1234", "W":0.5678}`, } for _, in := range inputs { t := T{} err := json.Unmarshal([]byte(in), &t) if err != nil { fmt.Println(err) continue } fmt.Printf("%s %s\n", t.V, t.W) }
Output: Rat.UnmarshalJSON: cannot convert []uint8([97 98]) to Rat 0 0 invalid character '}' looking for beginning of value 0 0 1 0 0.12345678 0 0.1234 0.5678
func (*Rat) Value ¶ added in v0.24.0
Value return the []byte value for database/sql, as defined in sql/driver.Valuer. It will return "0" if r is nil.
Example ¶
inputs := []interface{}{ nil, 0, 1.2345, "12345.6789_1234_5678_9", } for _, in := range inputs { out, _ := NewRat(in).Value() fmt.Printf("%s\n", out) }
Output: 0 0 1.2345 12345.67891234