Elasticsearch的安装及基本使用

目录隐藏

1 简介

2 安装

简介

Elasticsearch是一个高度可扩展的全文搜索和分析引擎，基于Apache Lucence（事实上，Lucence也是百度所采用的搜索引擎）构建，能够对大容量的数据进行接近实时的存储、搜索和分析操作。

安装

安装Java8

在安装Elasticsearch前，首先需要安装JAVA环境。
可以先执行以下命令查看是否安装：

java -version

如果没有安装需要先到oracle官网下载jdk-6u45-windows-x64.exe并安装。下载时会提示需要登录账号才能下载。

配置Java环境变量

在系统环境变量中，新建JAVA_HOME变量，值为：C:\Program Files\Java\jdk1.6.0_45：

在path这个系统环境变量中，添加：;%JAVA_HOME%\bin：

添加CLASSPATH变量，值为：.;%JAVA_HOME%\lib\dt.jar;%JAVA_HOME%\lib\tools.jar;

打开cmd命令行，输入javac -version，如果成功执行命令，则说明jdk安装成功。

安装Elasticsearch

下载地址：https://www.elastic.co/cn/downloads/elasticsearch

目前最新版是8.4.1

下载后解压压缩包，在config/elasticsearch.yml末尾添加：

ingest.geoip.downloader.enabled: false

然后修改如下配置：

network.host: 127.0.0.1
http.port: 9200

双击bin/elasticsearch.bat运行后，报错，再次打开elasticsearch.yml，修改：

xpack.security.enabled: false

双击elasticsearch.bat运行，浏览器访问127.0.0.1:9200,看到以下界面则说明安装成功：

基本使用

创建索引

PUT http://127.0.0.1:9200/<index>

<index>：索引的名称
注：也可以不创建索引，后续在添加文档时，如果没有索引会自动创建。

删除索引

DELETE http://127.0.0.1:9200/<index>

添加/更新文档

POST http://127.0.0.1:9200/<index>/_doc/<_id>

<index>：索引的名称
<_id>：文档的唯一标识符

例如：

POST http://127.0.0.1:9200/movies/_doc/1
{"id":1,"title":"Kung Fu Panda","overview":"When the Valley of Peace is threatened, lazy Po the panda discovers his destiny as the \"chosen one\" and trains to become a kung fu hero, but transforming the unsleek slacker into a brave warrior won't be easy. It's up to Master Shifu and the Furious Five -- Tigress, Crane, Mantis, Viper and Monkey -- to give it a try.","genres":["Action","Adventure","Animation","Family","Comedy"],"poster":"https://image.tmdb.org/t/p/w500/wWt4JYXTg5Wr3xBW2phBrMKgp3x.jpg","release_date":1212537600}

可以使用Postman来操作：

批量添加文档

PUT http://127.0.0.1:9200/<index>/_bulk

例如：

PUT http://127.0.0.1:9200/movies/_bulk
{"index":{"_id":1}}
{"id":1,"title":"Kung Fu Panda","overview":"When the Valley of Peace is threatened, lazy Po the panda discovers his destiny as the \"chosen one\" and trains to become a kung fu hero, but transforming the unsleek slacker into a brave warrior won't be easy. It's up to Master Shifu and the Furious Five -- Tigress, Crane, Mantis, Viper and Monkey -- to give it a try.","genres":["Action","Adventure","Animation","Family","Comedy"],"poster":"https://image.tmdb.org/t/p/w500/wWt4JYXTg5Wr3xBW2phBrMKgp3x.jpg","release_date":1212537600}
{"index":{"_id":2}}
{"id":2,"title":"Batman","overview":"Batman has not been seen for ten years. A new breed of criminal ravages Gotham City, forcing 55-year-old Bruce Wayne back into the cape and cowl. But, does he still have what it takes to fight crime in a new era?","genres":["Action","Animation","Mystery"],"poster":"https://image.tmdb.org/t/p/w500/kkjTbwV1Xnj8wBL52PjOcXzTbnb.jpg","release_date":1345507200}

注：提交数据的最后一行有一行空行\n不能省略。

获取文档

GET http://127.0.0.1:9200/<index>/_doc/<_id>

例如：

GET http://127.0.0.1:9200/movies/_doc/1

删除文档

DELETE http://127.0.0.1:9200/<index>/_doc/<_id>

搜索文档

URL参数查询

GET http://127.0.0.1:9200/<index>/_search?q=<keyword>&sort=<field>:<direction>

q：使用q参数来运行查询参数搜索
<keyword>：查询字符串
sort：排序

例如：

GET http://127.0.0.1:9200/movies/_search?q=panda&sort=id:asc

url中的更多参数请查看官方文档

响应正文

搜素后得到如下响应正文：

{
    "took": 1,
    "timed_out": false,
    "_shards": {
        "total": 1,
        "successful": 1,
        "skipped": 0,
        "failed": 0
    },
    "hits": {
        "total": {
            "value": 1,
            "relation": "eq"
        },
        "max_score": null,
        "hits": [
            {
                "_index": "movies",
                "_id": "1",
                "_score": null,
                "_ignored": [
                    "overview.keyword"
                ],
                "_source": {
                    "id": 1,
                    "title": "Kung Fu Panda",
                    "overview": "When the Valley of Peace...",
                    "genres": [
                        "Action",
                        ...
                    ],
                    "poster": "https://image.xxx.jpg",
                    "release_date": 1212537600
                },
                "sort": [
                    1
                ]
            }
        ]
    }
}

took – 执行搜索的时间（以毫秒为单位）
timed_out – 搜索是否超时
_shards – 搜索了多少个分片，以及搜索成功/失败的分片数
hits – 搜索结果
hits.total – 符合搜索条件的文档总数
hits.hits – 搜索结果数组（默认为前10个文档）
hits.sort – 结果的排序键
hits._score – 文档的相关性，数字越高，文档越相关。

更多参数解释可以查看官方文档

DSL查询

Elasticsearch 提供了基于 JSON 的完整 Query DSL（Domain Specific Language）来定义查询。

term查询
查询ID为1的文档：

GET http://127.0.0.1:9200/movies/_search
{
    "query": {
        "term": {
            "id": 1
        }
    }
}

一般使用term检索非文本的精确值，例如商品价格、商品ID、登录账号等。

terms查询
相当于多个term检索, 类似于SQL中in关键字的用法, 即在某些给定的数据中检索:

GET http://127.0.0.1:9200/movies/_search
{
    "query": {
        "terms": {
            "title.keyword": [
                "Kung Fu Panda", "Batman"
            ]
        }
    }
}