使用 Go 语言与 Elasticsearch 实现高效搜索服务

使用 Go 语言与 Elasticsearch 实现高效搜索服务

什么是 Elasticsearch

• Elasticsearch 是一个基于 Apache Lucene 构建的分布式搜索引擎，能够存储、搜索和分析大量数据。它具有高可扩展性、快速的搜索速度，支持全文检索、多字段查询和近实时数据分析。Elasticsearch 常与 Logstash 和 Kibana 一起构成 ELK 技术栈，广泛应用于日志分析、推荐系统和全文检索等领域。

安装 Elasticsearch

• 下载并安装：你可以从 Elasticsearch 官网 下载适用于你操作系统的版本。
• 启动 Elasticsearch：

./bin/elasticsearch

• 验证安装：可以通过访问以下地址来验证 Elasticsearch 是否启动成功

http://localhost:9200/

Go 语言与 Elasticsearch 集成

• 我们将使用 olivere/elastic 库，这是一个流行的 Go 语言客户端库，能够方便地与 Elasticsearch 交互。以下是实现的步骤：

安装 olivere/elastic 库

go get github.com/olivere/elastic/v7

• 这个库支持 Elasticsearch 7.x 版本的所有功能，确保你的 Elasticsearch 版本兼容。

连接 es

import ("github.com/olivere/elastic/v7""testing"
)func TestClient(t *testing.T) {client, err := elastic.NewClient(elastic.SetSniff(false),elastic.SetURL("http://localhost:9200"),)if err != nil {panic(err)}t.Log(client)
}

go 操作索引

1. 创建索引

import ("github.com/olivere/elastic/v7""testing"
)func TestIndexMapping(t *testing.T) {mapping := `{"settings" : {"number_of_shards" : 1,"number_of_replicas" : 1},"mappings":{"properties":{"nickname":{"type":"text"},"username":{"type":"keyword"},"id":{"type":"integer"},"create_at":{"type":"date","format":"[YYYY-MM-dd HH:mm:ss]"}}}}`//t.Log(mapping)client, err := elastic.NewClient(elastic.SetSniff(false),elastic.SetURL("http://localhost:9200"),)if err != nil {panic(err)}do, err1 := client.CreateIndex("user").BodyString(mapping).Do(context.Background())if err1 != nil {t.Logf("创建mapping失败%s", err1.Error())return}t.Log("创建mapping成功", do)}

2. 判断索引是否存在

import ("github.com/olivere/elastic/v7""testing"
)func TestIndexExists(t *testing.T) {client, err := elastic.NewClient(elastic.SetSniff(false),elastic.SetURL("http://localhost:9200"),)if err != nil {panic(err)}do, err1 := client.IndexExists("user").Do(context.Background())if err1 != nil {panic(err1)}t.Log(do)
}

3. 删除索引

import ("github.com/olivere/elastic/v7""testing"
)func TestDeleteIndex(t *testing.T) {client, err := elastic.NewClient(elastic.SetURL("http://localhost:9200"),elastic.SetSniff(false),)if err != nil {panic(err)}do, err1 := client.DeleteIndex("user").Do(context.Background())if err1 != nil {panic(err1)}t.Log(do)}

go 操作文档

1. 添加文档

import ("github.com/olivere/elastic/v7""testing"
)type User struct {ID       uint      `json:"id"`Username string    `json:"username"`Nickname string    `json:"nickname"`CreateAt time.Time `json:"create_at"`
}func TestCreateDoc(t *testing.T) {user := &User{ID:       1,Username: "晓智科技",Nickname: "晓智",CreateAt: time.Now(),}client, err := elastic.NewClient(elastic.SetSniff(false),elastic.SetURL("http://localhost:9200"),)if err != nil {panic(err)}do, err1 := client.Index().Index("user").BodyJson(user).Do(context.Background())if err1 != nil {panic(err1)}t.Log(do)}

2. 根据 id 删除

import ("github.com/olivere/elastic/v7""testing"
)func TestDeleteDoc(t *testing.T) {client, err := elastic.NewClient(elastic.SetSniff(false),elastic.SetURL("http://localhost:9200"),)if err != nil {panic(err)}deleteId := "7VKzh4sBP_UzlGBnhzrD"do, err1 := client.Delete().Index("user").Id(deleteId).Refresh("true").Do(context.Background())if err1 != nil {panic(err1)}t.Log(do)}

3. 根据 id 批量删除

import ("github.com/olivere/elastic/v7""testing"
)
func TestBatchDeleteDocById(t *testing.T) {list := []string{"91LDh4sBP_UzlGBnwzpC", "9VLDh4sBP_UzlGBntjr8"}client, err := elastic.NewClient(elastic.SetSniff(false),elastic.SetURL("http://localhost:9200"),)if err != nil {panic(err)}bulk := client.Bulk().Index("user").Refresh("true")for _, s := range list {req := elastic.NewBulkDeleteRequest().Id(s)bulk.Add(req)}do, err1 := bulk.Do(context.Background())if err1 != nil {panic(err1)}t.Log(do.Succeeded())
}

4. 批量添加

import ("github.com/olivere/elastic/v7""testing"
)func TestBatchCreate(t *testing.T) {list := []User{{ID:       12,Username: "张三",Nickname: "三",CreateAt: time.Now(),},{ID:       13,Username: "李四",Nickname: "阿四",CreateAt: time.Now(),},}client, err := elastic.NewClient(elastic.SetSniff(false),elastic.SetURL("http://localhost:9200"),)if err != nil {panic(err)}bulk := client.Bulk().Index("user").Refresh("true")for _, user := range list {doc := elastic.NewBulkCreateRequest().Doc(&user)bulk.Add(doc)}do, err1 := bulk.Do(context.Background())if err1 != nil {panic(err1)}t.Log(do.Created())
}

文档查询

1. 分页查询

import ("github.com/olivere/elastic/v7""testing"
)// 查询文档
func TestFindDoc(t *testing.T) {client, err := elastic.NewClient(elastic.SetSniff(false),elastic.SetURL("http://localhost:9200"),)if err != nil {panic(err)}query := elastic.NewBoolQuery()res, err1 := client.Search("user").Query(query).From(0).Size(10).Do(context.Background())if err1 != nil {panic(err1)}count := res.Hits.TotalHits.Valuet.Log(count)for _, value := range res.Hits.Hits {t.Log(string(fmt.Sprintf("%d", value.Score)))}}

2. 精确查询

import ("github.com/olivere/elastic/v7""testing"
)func TestFindTermDoc(t *testing.T) {client, err := elastic.NewClient(elastic.SetSniff(false),elastic.SetURL("http://localhost:9200"),)if err != nil {panic(err)}query := elastic.NewTermQuery("nickname", "晓智科技")res, err1 := client.Search("user").Query(query).From(0).Size(10).Do(context.Background())if err1 != nil {panic(err1)}count := res.Hits.TotalHits.Valuet.Log(count)for _, val := range res.Hits.Hits {t.Log(val.Source)}}

3. 模糊查询

import ("github.com/olivere/elastic/v7""testing"
)// 模湖查询
func TestFinMathDoc(t *testing.T) {client, err := elastic.NewClient(elastic.SetSniff(false),elastic.SetURL("http://localhost:9200"),)if err != nil {panic(err)}query := elastic.NewMatchQuery("desc", "It从业人员")res, err1 := client.Search("user").Query(query).From(0).Size(10).Do(context.Background())if err1 != nil {panic(err1)}count := res.Hits.TotalHits.Valuet.Log(count)for _, val := range res.Hits.Hits {t.Log(string(val.Source))}
}

4. 文档更新

import ("github.com/olivere/elastic/v7""testing"
)// 文档更新
func TestUpdateDoc(t *testing.T) {client, err := elastic.NewClient(elastic.SetSniff(false), elastic.SetURL("http://localhost:9200"))if err != nil {panic(err)}updateId := "8VLDh4sBP_UzlGBnlTqW"do, err1 := client.Update().Index("user").Id(updateId).Doc(map[string]any{"username": "晓晓智"}).Do(context.Background())if err1 != nil {panic(err1)}t.Log(do)
}

使用异步请求

• 为了提高响应速度，可以使用异步请求处理搜索和索引操作。异步请求不会阻塞主线程，可以提高吞吐量。

client.Index().Index("products").BodyJson(product).DoAsync(context.Background(), func(response *elastic.IndexResponse, err error) {if err != nil {log.Printf("Error indexing document asynchronously: %s", err)} else {fmt.Printf("Asynchronous indexing completed for document %s\n", response.Id)}})

相关链接

演示地址
获取更多
源码地址