go基于阿里云实现中文分词

API介绍

由于阿里云服务SDK方式调用只提供了JAVA和Python,用go就只能使用普通HTTP请求调用

阿里云NLP基础服务

API概览

获取AccessKey
中文分词(基础版)API
公共参数每个接口都需要使用的请求参数(公共参数概览);公共参数大多参数都给了对应的取值.唯一一个Signature(签名结果串)比较复杂(签名机制),其他的参数全部都拼接到地址后面就可以了

代码

package main

import (
	"crypto/hmac"
	"crypto/sha1"
	"encoding/base64"
	"encoding/json"
	"fmt"
	"github.com/satori/go.uuid"
	"io/ioutil"
	"net/http"
	url2 "net/url"
	"time"
)

func main() {
	nlp("今天，股市沸腾！发生了啥？")
}

var AccessKeyID = ""
var AccessKeySecret = ""

func nlp(text string) {
	if len(text) > 1024 {
		panic("text Max Length 1024")
	}
	duration, err := time.ParseDuration("-8h")
	if err != nil {
		panic(err)
	}
	timeStamp := time.Now().Add(duration).Format("2006-01-02T15:04:05Z")
	signatureNonce := uuid.NewV4().String()
	params := url2.Values{}
	params.Add("AccessKeyId", AccessKeyID)
	params.Add("Action", "GetWsChGeneral")
	params.Add("Format", "JSON")
	params.Add("ServiceCode", "alinlp")
	params.Add("SignatureMethod", "HMAC-SHA1")
	params.Add("SignatureNonce", signatureNonce)
	params.Add("SignatureVersion", "1.0")
	params.Add("Text", text)
	params.Add("Timestamp", timeStamp)
	params.Add("TokenizerId", "GENERAL_CHN")
	params.Add("Version", "2020-06-29")
	// 0：大粒度，1：中粒度（默认），2：小粒度
	params.Add("OutType", "1")

	// 将所有请求参数进行URL编码
	stringToSign := url2.QueryEscape(params.Encode())
	// 构造成待签名的字符串
	stringToSign = http.MethodGet + "&" + url2.QueryEscape("/") + "&" + stringToSign
	secret := AccessKeySecret + "&"

	// 计算签名值
	hash := hmac.New(sha1.New, []byte(secret))
	hash.Write([]byte(stringToSign))
	signature := base64.StdEncoding.EncodeToString(hash.Sum(nil))
	params.Add("Signature", signature)

	// 请求地址
	url := "https://alinlp.cn-hangzhou.aliyuncs.com/"
	uri, err := url2.ParseRequestURI(url)
	if err != nil {
		panic(err)
	}
	uri.RawQuery = params.Encode()

	//发送请求
	url = uri.String()
	resp, err := http.Get(url)
	if err != nil {
		panic(err)
	}
	defer resp.Body.Close()
	bytes, err := ioutil.ReadAll(resp.Body)
	if err != nil {
		panic(err)
	}

	var result RequestResult
	err = json.Unmarshal(bytes, &result)
	if err != nil {
		panic(err)
	}
	var data RequestResultData
	err = json.Unmarshal([]byte(result.Data), &data)
	if err != nil {
		panic(err)
	}
	bytes, err = json.Marshal(data)
	if err != nil {
		panic(err)
	}
	fmt.Println(string(bytes))
}

type RequestResult struct {
	RequestId string `json:"RequestId"`
	Data      string `json:"Data"`
}

type RequestResultData struct {
	Result  []DataResult `json:"result"`
	Success bool         `json:"success"`
}
type DataResult struct {
	ID   string   `json:"id"`
	Word string   `json:"word"`
	Tags []string `json:"tags"`
}

{
    "result": [
        {
            "id": "0",
            "word": "今天",
            "tags": [
                "基本词-中文",
                "机构-机构特指",
                "文体娱乐类-报纸杂志类",
                "文体娱乐类-戏剧歌曲类",
                "文体娱乐类-影视类",
                "文体娱乐类-书文课程类"
            ]
        },
        {
            "id": "1",
            "word": "，",
            "tags": []
        },
        {
            "id": "2",
            "word": "股市",
            "tags": [
                "基本词-中文",
                "产品类型修饰词"
            ]
        },
        {
            "id": "3",
            "word": "沸腾",
            "tags": [
                "基本词-中文",
                "产品-品牌"
            ]
        },
        {
            "id": "4",
            "word": "！",
            "tags": []
        },
        {
            "id": "5",
            "word": "发生",
            "tags": [
                "基本词-中文"
            ]
        },
        {
            "id": "6",
            "word": "了",
            "tags": [
                "基本词-中文",
                "文体娱乐类-flash作品"
            ]
        },
        {
            "id": "7",
            "word": "啥",
            "tags": [
                "基本词-中文"
            ]
        },
        {
            "id": "8",
            "word": "？",
            "tags": []
        }
    ],
    "success": true
}