golang解析excel、csv编码格式

开箱即用的工具

前言

utf8gbkGB18030

excel和csv中有多种编码格式，也是比较头疼的。这里提供多种解析方式。开箱即用的工具

源码

// 转为utf8格式
ConvertToString("你的字符")

package utils

import (
	"github.com/axgle/mahonia"
	"golang.org/x/text/encoding/simplifiedchinese"
)

const (
	GBK     string = "GBK"
	UTF8    string = "UTF8"
	UNKNOWN string = "UNKNOWN"
)

func ConvertToStringByCode(src string, srcCode string, tagCode string) string {
	srcCoder := mahonia.NewDecoder(srcCode)
	srcResult := srcCoder.ConvertString(src)
	tagCoder := mahonia.NewDecoder(tagCode)
	_, cdata, _ := tagCoder.Translate([]byte(srcResult), true)
	result := string(cdata)
	return result
}

func ConvertToStringGbkToUtf8(src string) string {
	if isUtf8([]byte(src)) {
		return src
	}
	var srcCode = "gbk"
	var tagCode = "utf-8"
	srcCoder := mahonia.NewDecoder(srcCode)
	srcResult := srcCoder.ConvertString(src)
	tagCoder := mahonia.NewDecoder(tagCode)
	_, cdata, _ := tagCoder.Translate([]byte(srcResult), true)
	result := string(cdata)
	return result
}

func ConvertToString(src string) string {
	if isUtf8([]byte(src)) {
		return src
	}
	if isGBK([]byte(src)) {
		utf8Data, err := simplifiedchinese.GBK.NewDecoder().Bytes([]byte(src))
		if err != nil {
			return src
		}
		if isUtf8(utf8Data) {
			return string(utf8Data)
		} else {
			return ConvertToStringGbkToUtf8(string(utf8Data))
		}
	}
	sink, err := simplifiedchinese.GB18030.NewDecoder().Bytes([]byte(src))
	if err != nil {
		return src
	}
	return string(sink)
}

func isGBK(data []byte) bool {
	length := len(data)
	var i int = 0
	for i < length {
		if data[i] <= 0x7f {
			//编码0~127,只有一个字节的编码，兼容ASCII码
			i++
			continue
		} else {
			//大于127的使用双字节编码，落在gbk编码范围内的字符
			if data[i] >= 0x81 &&
				data[i] <= 0xfe &&
				data[i+1] >= 0x40 &&
				data[i+1] <= 0xfe &&
				data[i+1] != 0xf7 {
				i += 2
				continue
			} else {
				return false
			}
		}
	}
	return true
}

func preNUm(data byte) int {
	var mask byte = 0x80
	var num int = 0
	//8bit中首个0bit前有多少个1bits
	for i := 0; i < 8; i++ {
		if (data & mask) == mask {
			num++
			mask = mask >> 1
		} else {
			break
		}
	}
	return num
}

func isUtf8(data []byte) bool {
	i := 0
	for i < len(data) {
		if (data[i] & 0x80) == 0x00 {
			// 0XXX_XXXX
			i++
			continue
		} else if num := preNUm(data[i]); num > 2 {
			// 110X_XXXX 10XX_XXXX
			// 1110_XXXX 10XX_XXXX 10XX_XXXX
			// 1111_0XXX 10XX_XXXX 10XX_XXXX 10XX_XXXX
			// 1111_10XX 10XX_XXXX 10XX_XXXX 10XX_XXXX 10XX_XXXX
			// 1111_110X 10XX_XXXX 10XX_XXXX 10XX_XXXX 10XX_XXXX 10XX_XXXX
			// preNUm() 返回首个字节的8个bits中首个0bit前面1bit的个数，该数量也是该字符所使用的字节数
			i++
			for j := 0; j < num-1; j++ {
				//判断后面的 num - 1 个字节是不是都是10开头
				if (data[i] & 0xc0) != 0x80 {
					return false
				}
				i++
			}
		} else {
			//其他情况说明不是utf-8
			return false
		}
	}
	return true
}

func GetStrCoding(data []byte) string {
	if isUtf8(data) == true {
		return UTF8
	} else if isGBK(data) == true {
		return GBK
	} else {
		return UNKNOWN
	}
}