CODEWITHSUNDEEP
go
Andrew Sullivan
Andrew Sullivan

Reputation: 67

How can I convert from big.Rat to a string without specifying precision?

I have an endpoint implemented in golang that can receive amounts with different precision length, ie:

"123"
"123.12"
"123.123123"

I'm using big.Rat internally to deal with these numbers as follows:

import (
    "encoding/json"
    "math/big"
)

type MyStruct struct {
    Amount   big.Rat `json:"amount"`
}

func (mystr *MyStruct) MarshalJSON() ([]byte, error) {
    type Alias MyStruct

    return json.Marshal(&struct {
        Amount json.Number `json:"amount"`
        *Alias
    }{
        Amount: json.Number(mystr.Amount.FloatString(2)),
        Alias:  (*Alias)(mystr),
    })
}

func (mystr *MyStruct) UnmarshalJSON(data []byte) error {
    type Alias MyStruct

    aux := &struct {
        Amount json.Number `json:"amount"`
        *Alias
    }{
        Alias: (*Alias)(mystr),
    }

    if err := json.Unmarshal(data, &aux); err != nil {
        return err
    }

    mystr.Amount = mystr.toRat(aux.Amount.String())

    return nil
}

func (mystr *MyStruct) toRat(val string) big.Rat {
    ratAmount := new(big.Rat)
    ratAmount.SetString(val)

    return *ratAmount
}

My problem & question is related to the Marshal method, in particular with this line:

Amount: json.Number(mystr.Amount.FloatString(2)),

Because if the amount in the json has a number with more than two decimal places a rounding takes place and I don't want that, I just want to mantain exactly the same number that was received when I did the Unmarshal method.

This is an example of rounding: https://play.golang.org/p/U6oi_aGc8lE

Is there a way to convert from big.Rat to string without defining the precision?

Thank you very much in advance!

Upvotes: 2

Views: 3128

Answers (1)

Santiago
Santiago

Reputation: 1745

This should be help:

package main

import (
    "fmt"
    "math/big"
    "strconv"
)

func main() {
    n := new(big.Rat)

    n.SetString("34.999999")
    x,_ := n.Float64()
    fmt.Println("Result: "+strconv.FormatFloat(x, 'f', -1, 64))
}

Upvotes: 1

Related Questions