1. 程式人生 > >base58編碼原理和實現(golang)

base58編碼原理和實現(golang)

在看EOS原始碼時,接觸到base58編碼。和通常base64編碼一樣,base58編碼的作用也是將非可視字元視覺化(ASCII化)。但不同的是base58編碼去掉了幾個看起來會產生歧義的字元,如 0 (零), O (大寫字母O), I (大寫的字母i) and l (小寫的字母L) ,和幾個影響雙擊選擇的字元,如/, +。結果字符集正好58個字元(包括9個數字,24個大寫字母,25個小寫字母)。而且因為58 不是2的整次冪,所以沒有使用類似base64編碼中使用直接擷取3個字元轉4個字元(3*8=4*6 , 2的6次方剛好64)的方法進行轉換,而是採用我們數學上經常使用的進位制轉換方法——輾轉相除法(本質上,base64編碼是64進位制,base58是58進位制

)。看下base58的編碼表:


也就是字元1代表0,字元2代表1,字元3代表2...字元z代表57。然後回一下輾轉相除法。

如要將1234轉換為58進位制;

第一步:1234除於58,商21,餘數為16,查表得H

第二步:21除於58,商0,餘數為21,查表得N

所以得到base58編碼為:NH

如果待轉換的數前面有0怎麼辦?直接附加編碼1來代表,有多少個就附加多少個(編碼表中1代表0)。現在我們看下go語言中的實現:

// Copyright (c) 2015 The btcsuite developers
// Use of this source code is governed by an ISC
// license that can be found in the LICENSE file.

// AUTOGENERATED by genalphabet.go; do not edit.

package base58

import (
	"math/big"
)
const (
	// alphabet is the modified base58 alphabet used by Bitcoin.
	alphabet = "123456789ABCDEFGHJKLMNPQRSTUVWXYZabcdefghijkmnopqrstuvwxyz"

	alphabetIdx0 = '1'
)

var  b58 = [256]byte{
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 0, 1, 2, 3, 4, 5, 6,
	7, 8, 255, 255, 255, 255, 255, 255,
	255, 9, 10, 11, 12, 13, 14, 15,
	16, 255, 17, 18, 19, 20, 21, 255,
	22, 23, 24, 25, 26, 27, 28, 29,
	30, 31, 32, 255, 255, 255, 255, 255,
	255, 33, 34, 35, 36, 37, 38, 39,
	40, 41, 42, 43, 255, 44, 45, 46,
	47, 48, 49, 50, 51, 52, 53, 54,
	55, 56, 57, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
	255, 255, 255, 255, 255, 255, 255, 255,
}
//go:generate go run genalphabet.go

var bigRadix = big.NewInt(58)
var bigZero = big.NewInt(0)

// Decode decodes a modified base58 string to a byte slice.
func Decode(b string) []byte {
	answer := big.NewInt(0)
	j := big.NewInt(1)

	scratch := new(big.Int)
	for i := len(b) - 1; i >= 0; i-- {
		//字元,ascii碼錶的簡版-->得到字元代表的值(0,1,2,..57)
		tmp := b58[b[i]]
		//出現不該出現的字元
		if tmp == 255 {
			return []byte("")
		}
		scratch.SetInt64(int64(tmp))

		//scratch = j*scratch
		scratch.Mul(j, scratch)

		answer.Add(answer, scratch)
		//每次進位都要乘上58
		j.Mul(j, bigRadix)
	}

	//得到大端的位元組序
	tmpval := answer.Bytes()

	var numZeros int
	for numZeros = 0; numZeros < len(b); numZeros++ {
		//得到高位0的位數
		if b[numZeros] != alphabetIdx0 {
			break
		}
	}
	//得到原來數字的長度
	flen := numZeros + len(tmpval)
	//構造一個新地存放結果的空間
	val := make([]byte, flen, flen)
	copy(val[numZeros:], tmpval)

	return val
}

// Encode encodes a byte slice to a modified base58 string.
func Encode(b []byte) string {
	x := new(big.Int)
	//將b解釋為大端儲存
	x.SetBytes(b)

	//Base58編碼可以表示的位元位數為Log258 {\displaystyle \approx } \approx5.858bit。經過Base58編碼的資料為原始的資料長度的1.37倍
	answer := make([]byte, 0, len(b)*136/100)

	for x.Cmp(bigZero) > 0 {
		mod := new(big.Int)
		//x除於58的餘數mod,並將商賦值給x
		x.DivMod(x, bigRadix, mod)
		answer = append(answer, alphabet[mod.Int64()])
	}

	// leading zero bytes
	//因為如果高位為0,0除任何數為0,可以直接設定為‘1’
	for _, i := range b {
		if i != 0 {
			break
		}
		answer = append(answer, alphabetIdx0)
	}

	// reverse
	//因為之前先附加低位的,後附加高位的,所以需要翻轉
	alen := len(answer)
	for i := 0; i < alen/2; i++ {
		answer[i], answer[alen-1-i] = answer[alen-1-i], answer[i]
	}

	return string(answer)
}