搜索与分析ElasticSearch之一：URL请求

ElasticSearch

一 启动ElasticSearch

点击elasticsearch bin目录下的第一个bat

浏览器输入网址 localhost:9200

二 图形化工具

有两种工具可选elasticsearch-head与kibana

2.1 elasticsearch-head

需要nodejs环境，在elasticsearch-head目录中打开cmd 输入

初次使用需要如下语句：

npm install -g grunt-cli
npm install

启动head：

grunt server

然后访问 http://localhost:9100/，并连接 http://localhost:9200/

如果不能连接，打开elasticsearch中的config/elasticsearch.yml，添加跨域允许

http.cors.enabled: true
http.cors.allow-origin: "*"

2.2 Kibana

下载对应Es版本的Kibana，然后双击bin目录下的kiban.bat
kibana改为中文: kibana.yml末尾增加

 i18n.locale: "zh-CN"

三 增删改查

3.1 增

增操作包含：1增加索引 & 2增加文档

3.1.1 一次性生成索引

PUT http://localhost:9200/blog
{
    "settings": {
        "number_of_shards":3,
        "number_of_replicas":1
    },
    "mappings":{
        "properties":{
            "type":{"type":"keyword"},
            "name":{"type":"text"},
            "country":{"type":"keyword"},
            "age":{"type":"integer"},
            "date":{
                "type":"data",
                "format":"yyyy-MM-dd HH:mm:ss || yyyy-MM-dd || epoch_millis"
            }
        }
    }
}

3.1.2 先生成空索引，然后在添加在设置mappings

（1）先新建索引

PUT http://ip:port/索引名称

PUT http://localhost:9200/blog
{
    "settings": {
        "number_of_shards":3,
        "number_of_replicas":1
    }
}

（2）在设置_mappings

post http://ip:port/索引名称/_mappings

POST localhost:9200/blog/_mappings
{
	"properties":{
		"id":{
            "type":"long",
            "store":true
		},
        "title":{
            "type":"text",
            "store":true,
            "index":true,
            "analyzer":"standard"
        },
        "content":{
            "type":"text",
            "store":true,
            "index":true,
            "analyzer":"standard"
        }
	}
}

3.1.3 向索引库中添加文档

POST http://ip:port/索引名称/_doc/文档ID（可以不填）

POST http://localhost:9200/blog/_doc/1
{
	"id":1,
    "title":"学习java的全文检索工具",
    "content":"Luence是某某某家伙写的，这个东西真的超级不好用，所有又有人写了ElasticSearch"
}

如果不填写ID，那么将自动生成ID

3.2 删

删除操作主要分：1 删除文档 & 2 删除索引

3.2.1 删除文档

delete http://ip:port/索引名称/_doc/_id

3.2.2 删除索引

delete http://ip:port/索引名称

3.3 改

注：更新\修改的原理是先删除后添加，因为Luence是这个原理

3.3.1 修改文档数据

post http://ip:port/索引名称/_update/_id

{
	"doc":{
		"title":"我去前面探探路"
	}
}

实例：修改索引blog中id为3的文档title为 ”我去前面探探路“

POST http://localhost:9200/blog/_update/3
{
    "doc":{
        "title":"我去前面探探路"
    }
}

3.3.2 修改索引

3.4 查

3.4.1 根据ID查询

GET http://ip:port/索引名称/_doc/_id

GET http://localhost:9200/blog/_doc/3

3.4.2 根据关键词查询

仅仅对输入的内容进行查询

post http://ip:port/索引名称/_search/
{
	"query":{
		"term":{
			"title":"java"
		}
	}
}

3.4.3 queryString查询

根据输入的内容进行分析后查询

post http://ip:port/索引名称/_search/
{
	"query":{
		"query_string":{
			"default_field":"title",
			"query":"java"
		}
	}
}

3.4.4 IK分词器

3.4.4.1分词器效果查看

注意：在7.X版本后分词器写法有更改如果在7.X使用以下写法：

get http://localhost:9200/_analyze?analyzer=standard&text=variously spelled wildcard or wild-card,alse known as at-large berth

返回结果报错为：

{
    "error": {
        "root_cause": [
            {
                "type": "parse_exception",
                "reason": "request body or source parameter is required"
            }
        ],
        "type": "parse_exception",
        "reason": "request body or source parameter is required"
    },
    "status": 400
}

正确写法：

get http://localhost:9200/_analyze
{
	"analyzer":"standard",
	"text":"variously spelled wildcard or wild-card,alse known as at-large berth"
}

3.4.4.2 安装IK插件

1> https://github.com/medcl/elasticsearch-analysis-ik 下载对应版本IK

下载的版本一定要与ES的版本对应，如果不是闲得慌，千万不要尝试最新版的ES和IK

2> 解压后放到plugins目录，重启即可！

3.4.4.3 重新构建以IK为基础的索引

put http://localhost:9200/blog
{
  "settings": {
    "number_of_shards": 3,
    "number_of_replicas": 1
  },
  "mappings": {
    "properties": {
      "id": {"type": "long", "store": "true","analyzer": "ik_max_word", "search_analyzer": "ik_smart"},
      "title": {"type": "text","store":"true","analyzer": "ik_max_word", "search_analyzer": "ik_smart"},
      "content": {"type": "text","store":"true","analyzer": "ik_max_word", "search_analyzer": "ik_smart"}
    }
  }
}

新建时报错：

{	"type": "access_control_exception",
	"reason": "access denied (\"java.io.FilePermission\" \"D:\\Program%20Files%20(x86)\\programer\\elasticsearch-7.4.0\\plugins\\elasticsearch-analysis-ik-7.4.0\\config\\IKAnalyzer.cfg.xml\" \"read\")"}

原因：elasticsearch安装路径存在空格

更改路径后继续报错：

{
    "error": {
        "root_cause": [
            {
                "type": "mapper_parsing_exception",
                "reason": "Mapping definition for [id] has unsupported parameters:  [analyzer : ik_smart]"
            }
        ],
        "type": "mapper_parsing_exception",
        "reason": "Failed to parse mapping [_doc]: Mapping definition for [id] has unsupported parameters:  [analyzer : ik_smart]",
        "caused_by": {
            "type": "mapper_parsing_exception",
            "reason": "Mapping definition for [id] has unsupported parameters:  [analyzer : ik_smart]"
        }
    },
    "status": 400
}

原因：ID不能用IK，删掉ID那一行的analyzer和search_analyzer即可

3.4.4.3 对查询结果高亮

post http://localhost:9200/blog/_search
{
    "query" : { "match" : { "content" : "校车" }},
    "highlight" : {
        "pre_tags" : ["<tag1>", "<tag2>"],
        "post_tags" : ["</tag1>", "</tag2>"],
        "fields" : {
            "content" : {}
        }
    }
}

结果：

"hits": [
    {
        "_index": "blog",
        "_type": "_doc",
        "_id": "2",
        "_score": 0.2876821,
        "_source": {
            "id": "2",
            "title": "外国人永久居留条例！",
            "content": "美国留给伊拉克的是个烂摊子吗！公安部：各地校车将享最高路权"
        },
        "highlight": {
            "content": [
                "公安部：各地<tag1>校车</tag1>将享最高路权"
            ]
        }
    }
]

四 评分机制

4.1 tfidf算法