Elasticsearch golang's sdk usage

Documentation first

The "elasticsearch authoritative guide" directly looks at the online version of the official website. It is relatively new. Those pdf versions on the Internet are all version 2.x. Many are not compatible with the
official API manual . You can choose the version.

Choice of golang sdk library

There are mainly the following two

• github.com/olivere/elastic Third-party development, each version has a corresponding sdk, and the documentation is also rich
• github.com/elastic/go-elasticsearch no comments
  , we ended up choosing olivere/elastic

The corresponding codes of restful api and olivere/elastic are as follows:

package main

import (
    "context"
    "encoding/json"
    "fmt"
    "gopkg.in/olivere/elastic.v5" //这里使用的是版本5，最新的是6，有改动
    "log"
    "os"
    "reflect"
)

var client *elastic.Client
var host = "http://47.96.18.118:9200/"

type Employee struct {
    FirstName string   `json:"first_name"`
    LastName  string   `json:"last_name"`
    Age       int      `json:"age"`
    About     string   `json:"about"`
    Interests []string `json:"interests"`
}

//初始化
func init() {
    errorlog := log.New(os.Stdout, "APP", log.LstdFlags)
    var err error
    client, err = elastic.NewClient(elastic.SetErrorLog(errorlog), elastic.SetURL(host))
    if err != nil {
        panic(err)
    }
    info, code, err := client.Ping(host).Do(context.Background())
    if err != nil {
        panic(err)
    }
    fmt.Printf("Elasticsearch returned with code %d and version %s\n", code, info.Version.Number)

    esversion, err := client.ElasticsearchVersion(host)
    if err != nil {
        panic(err)
    }
    fmt.Printf("Elasticsearch version %s\n", esversion)

}

/*下面是简单的CURD*/

//创建
func create() {

    //使用结构体
    e1 := Employee{"Jane", "Smith", 32, "I like to collect rock albums", []string{"music"}}
    put1, err := client.Index().
        Index("megacorp").
        Type("employee").
        Id("1").
        BodyJson(e1).
        Do(context.Background())
    if err != nil {
        panic(err)
    }
    fmt.Printf("Indexed tweet %s to index s%s, type %s\n", put1.Id, put1.Index, put1.Type)

    //使用字符串
    e2 := `{"first_name":"John","last_name":"Smith","age":25,"about":"I love to go rock climbing","interests":["sports","music"]}`
    put2, err := client.Index().
        Index("megacorp").
        Type("employee").
        Id("2").
        BodyJson(e2).
        Do(context.Background())
    if err != nil {
        panic(err)
    }
    fmt.Printf("Indexed tweet %s to index s%s, type %s\n", put2.Id, put2.Index, put2.Type)

    e3 := `{"first_name":"Douglas","last_name":"Fir","age":35,"about":"I like to build cabinets","interests":["forestry"]}`
    put3, err := client.Index().
        Index("megacorp").
        Type("employee").
        Id("3").
        BodyJson(e3).
        Do(context.Background())
    if err != nil {
        panic(err)
    }
    fmt.Printf("Indexed tweet %s to index s%s, type %s\n", put3.Id, put3.Index, put3.Type)

}

//删除
func delete() {

    res, err := client.Delete().Index("megacorp").
        Type("employee").
        Id("1").
        Do(context.Background())
    if err != nil {
        println(err.Error())
        return
    }
    fmt.Printf("delete result %s\n", res.Result)
}

//修改
func update() {
    res, err := client.Update().
        Index("megacorp").
        Type("employee").
        Id("2").
        Doc(map[string]interface{}{"age": 88}).
        Do(context.Background())
    if err != nil {
        println(err.Error())
    }
    fmt.Printf("update age %s\n", res.Result)

}

//查找
func gets() {
    //通过id查找
    get1, err := client.Get().Index("megacorp").Type("employee").Id("2").Do(context.Background())
    if err != nil {
        panic(err)
    }
    if get1.Found {
        fmt.Printf("Got document %s in version %d from index %s, type %s\n", get1.Id, get1.Version, get1.Index, get1.Type)
    }
}

//搜索
func query() {
    var res *elastic.SearchResult
    var err error
    //取所有
    res, err = client.Search("megacorp").Type("employee").Do(context.Background())
    printEmployee(res, err)

    //字段相等
    q := elastic.NewQueryStringQuery("last_name:Smith")
    res, err = client.Search("megacorp").Type("employee").Query(q).Do(context.Background())
    if err != nil {
        println(err.Error())
    }
    printEmployee(res, err)

    if res.Hits.TotalHits > 0 {
        fmt.Printf("Found a total of %d Employee \n", res.Hits.TotalHits)

        for _, hit := range res.Hits.Hits {

            var t Employee
            err := json.Unmarshal(*hit.Source, &t) //另外一种取数据的方法
            if err != nil {
                fmt.Println("Deserialization failed")
            }

            fmt.Printf("Employee name %s : %s\n", t.FirstName, t.LastName)
        }
    } else {
        fmt.Printf("Found no Employee \n")
    }

    //条件查询
    //年龄大于30岁的
    boolQ := elastic.NewBoolQuery()
    boolQ.Must(elastic.NewMatchQuery("last_name", "smith"))
    boolQ.Filter(elastic.NewRangeQuery("age").Gt(30))
    res, err = client.Search("megacorp").Type("employee").Query(q).Do(context.Background())
    printEmployee(res, err)

    //短语搜索 搜索about字段中有 rock climbing
    matchPhraseQuery := elastic.NewMatchPhraseQuery("about", "rock climbing")
    res, err = client.Search("megacorp").Type("employee").Query(matchPhraseQuery).Do(context.Background())
    printEmployee(res, err)

    //分析 interests
    aggs := elastic.NewTermsAggregation().Field("interests")
    res, err = client.Search("megacorp").Type("employee").Aggregation("all_interests", aggs).Do(context.Background())
    printEmployee(res, err)

}

//打印查询到的Employee
func printEmployee(res *elastic.SearchResult, err error) {
    if err != nil {
        print(err.Error())
        return
    }
    var typ Employee
    for _, item := range res.Each(reflect.TypeOf(typ)) { //从搜索结果中取数据的方法
        t := item.(Employee)
        fmt.Printf("%#v\n", t)
    }
}

func main() {
    create()
    delete()
    update()
    gets()
    query()
}

Auxiliary tool

• ElasticHD single-file executable file, and supports sql to es query
• After mobz/elasticsearch-head es 5.0, plug-in installation is not supported. Instead, you need to run the http server through node, or through the Google browser plug-in,
  and finally choose ElasticHD