下载地址：https://www.elastic.co/cn/downloads/past-releases#elasticsearch \
IK分词器
下载地址
Releases · medcl/elasticsearch-analysis-ik · GitHub
索引库操作
fieldType详见：Field data types | Elasticsearch Guide [8.3] | Elastic
#创建索引库
PUT person
{
"settings": {
"number_of_shards": 5,
"number_of_replicas": 2
},
"mappings": {
"properties": {
"name":{
"type": "text"
},
"age":{
"type": "integer"
},
"sex":{
"type": "integer"
},
"birth":{
"type": "date"
}
}
}
}
#查看索引库的信息
GET person
#删除索引
DELETE person
7.2：(document)文档操作
7.2.1：添加操作
#添加文档
PUT person/_doc/1
{
"name":"jack",
"age":18,
"birth":"2018-11-11",
"address":"美国"
}
#不指定id自动生成id
POST person/_doc
{
"name":"rose",
"age":18,
"birth":"2018-11-11",
"address":"美国"
}
#批量添加
#注意：json不要换行
PUT _bulk
{"index":{"_index":"person","_id":"4"}}
{"doc":{"name":"jack2"}}
{"index":{"_index":"person","_id":"3"}}
{"doc":{"name":"jack3"}}
GET /person/_search
{
"query": {
"match_all": {}
}
}
7.2.2：修改文档
#根据id直接覆盖修改
POST /person/_doc/3
{
"email":"11@qq.com"
}
#指定修改某个field
POST person/_update/1
{
"doc":{
"name":"rose"
}
}
#批量修改
POST _bulk
{"update":{"_index":"person","_id":"1"}}
{"doc":{"name":"张三"}}
{"update":{"_index":"person","_id":"2"}}
{"doc":{"name":"张三1"}}
{"update":{"_index":"person","_id":"3"}}
{"doc":{"name":"张三2"}}
7.2.3：删除操作
#删除文档
DELETE person/_doc/4
#批量删除
PUT _bulk
{"delete":{"_index":"person","_id":"1"}}
{"delete":{"_index":"person","_id":"2"}}
八：查询操作
官方提供的测试数据地址：https://github.com/elastic/elasticsearch/blob/master/docs/src/test/resources/accounts.json
#一定要指定索引默认使用ik，要不然默认使用standard
#ik分词器 两种分词效果 ik_max_word、ik_smart
DELETE es_user
PUT es_user
{
"settings" : {
"index" : {
"analysis.analyzer.default.type": "ik_max_word"
}
}
}
PUT es_user/_bulk
{"index":{"_id":"1"}}
{"account_number":1,"balance":39225,"firstname":"Amber","lastname":"Duke","age":32,"gender":0,"address":"湖北省武汉市千锋教育，武汉金融港","job":"开发工程师","email":"amberduke@pyrami.com","city":"Brogan","state":"IL"}
{"index":{"_id":"6"}}
{"account_number":6,"balance":5686,"firstname":"Hattie","lastname":"Bond","age":36,"gender":1,"address":"湖北省武汉市千锋教育","job":"工程师","email":"hattiebond@netagy.com","city":"Dante","state":"TN"}
{"index":{"_id":"13"}}
{"account_number":13,"balance":32838,"firstname":"Nanette","lastname":"Bates","age":28,"gender":1,"address":"教育","job":"JAVA开发工程师","email":"nanettebates@quility.com","city":"Nogal","state":"VA"}
{"index":{"_id":"18"}}
{"account_number":18,"balance":4180,"firstname":"Dale","lastname":"Adams","age":33,"gender":1,"address":"中华人名共和国","job":"JAVA开发","email":"湖北省武汉市@boink.com","city":"Orick","state":"MD"}
{"index":{"_id":"20"}}
{"account_number":20,"balance":16418,"firstname":"Elinor","lastname":"Ratliff","age":36,"gender":1,"address":"山东省济南市","job":"产品经理","email":"高富帅@scentric.com","city":"Ribera","state":"WA"}
{"index":{"_id":"25"}}
{"account_number":25,"balance":40540,"firstname":"Virginia","lastname":"Ayala","age":19,"gender":0,"address":"湖南省常德市","job":"产品经理","email":"白富美@filodyne.com","city":"Nicholson","state":"PA"}
{"index":{"_id":"32"}}
{"account_number":32,"balance":48086,"firstname":"Dillard","lastname":"Mcpherson","age":34,"gender":0,"address":"河南省洛阳市","job":"产品经理","email":"北京@quailcom.com","city":"Veguita","state":"IN"}
{"index":{"_id":"37"}}
{"account_number":37,"balance":18612,"firstname":"Mcgee","lastname":"Mooney","age":18,"gender":0,"address":"河南省郑州市","job":"产品经理","email":"mcgeemooney@reversus.com","city":"Tooleville","state":"OK"}
{"index":{"_id":"44"}}
{"account_number":44,"balance":34487,"firstname":"Aurelia","lastname":"Harding","age":17,"gender":0,"address":"山东省青岛市","job":"项目经理","email":"aureliaharding@orbalix.com","city":"Yardville","state":"DE"}
{"index":{"_id":"49"}}
{"account_number":49,"balance":29104,"firstname":"Fulton","lastname":"Holt","age":23,"gender":0,"address":"山东省威海市","job":"项目经理","email":"千锋教育@anocha.com","city":"Sunriver","state":"RI"}
{"index":{"_id":"51"}}
{"account_number":51,"balance":14097,"firstname":"Burton","lastname":"Meyers","age":31,"gender":1,"address":"河南省开封市","job":"前端工程师","email":"burtonmeyers@bezal.com","city":"Jacksonburg","state":"MO"}
{"index":{"_id":"56"}}
{"account_number":56,"balance":14992,"firstname":"Josie","lastname":"Nelson","age":32,"gender":1,"address":"陕西省西安市","job":"前端工程师","email":"josienelson@emtrac.com","city":"Sunnyside","state":"UT"}
{"index":{"_id":"63"}}
{"account_number":63,"balance":6077,"firstname":"Hughes","lastname":"Owens","age":30,"gender":0,"address":"中国香港","job":"前端工程师","email":"hughesowens@valpreal.com","city":"Guilford","state":"KS"}
{"index":{"_id":"68"}}
{"account_number":68,"balance":44214,"firstname":"Hall","lastname":"Key","age":25,"gender":1,"address":"中国台湾","job":"前端工程师","email":"hallkey@eventex.com","city":"Shawmut","state":"CA"}
8.1：id和ids
GET es_user/_doc/1
GET es_user/_search
{
"query": {
"ids": {"values": [1,2,13]}
}
}
8.2：match查询【重要】
#match_all,查询所有
GET emp/_search
{
"query": {
"match_all": {}
}
}
GET _analyze
{
"analyzer": "ik_max_word",
"text": "湖北省武汉市千锋教育"
}
#match
#1:将《湖北省》进行中文分词《湖北》《湖北省》《省》
#2：将分词结果逐一匹配词条
GET emp/_search
{
"query": {
"match": {
"address": "湖北省"
}
}
}
#bool match
#query的值不会分词，如果是and需要每个词条都要命中某条文档
#如果是or，只需要其中一个词条命中某个文档即可
GET emp/_search
{
"query": {
"match": {
"address": {
"query": "湖北 我爱武汉",
"operator": "and"
}
}
}
}
#query的值是先分词，再匹配词条
GET emp/_search
{
"query": {
"match": {
"address": {
"query": "湖北 我爱武汉",
"operator": "or"
}
}
}
}
#multi_match多域查询
#query的值会分词，然后在多个域中匹配词条，只要其中一个域能匹配即可
GET emp/_search
{
"query": {
"multi_match": {
"query": "我爱湖北",
"fields": ["address","email"]
}
}
}
关键字检索，可以使用match进行检索，因为match是先分词再匹配词条
8.3：term查询【重要】
#term
#不分词直接匹配词条
GET emp/_search
{
"query": {
"term": {
"address": {
"value": "武汉好地方"
}
}
}
}
按商品分类或者品牌这类检索，可以使用term检索
8.4：prefix查询
#prefix
#词条以指定的value为前缀的
GET emp/_search
{
"query": {
"prefix": {
"address": {
"value": "武"
}
}
}
}
8.5：wildcard查询
不分词通配符的方式匹配词条
#wildcard
GET emp/_search
{
"query": {
"wildcard": {
"address": {
"value": "武汉*"
}
}
}
}
GET emp/_search
{
"query": {
"wildcard": {
"address": {
"value": "武汉??"
}
}
}
}
8.6：range查询【重要】
POST z/_search
{
"query": {
"range": {
"age": {
"gte": 32,
"lte": 36
}
}
}
}
8.7：分页查询【重要】
from+size
POST z/_search
{
"from": 2,
"size": 2,
"query": {
"match_all": {
}
}
}
8.8：复合查询【重要】
must:求交集
must not：取反
should：并集
must 多个查询单元必须同时匹配
GET z/_search
{
"query": {
"bool": {
"must": [
{
"term": {
"address": {
"value": "湖北省"
}
}
},
{
"range": {
"balance": {
"gte": 5600,
"lte": 6000
}
}
}
]
}
}
}
must not 多个查询单元必须都不匹配
GET z/_search
{
"query": {
"bool": {
"must_not": [
{
"match": {
"address": "湖北湖南"
}
},
{
"range": {
"age": {
"gte": 30,
"lte": 36
}
}
}
]
}
}
}
should:多个查询单元满足其中一个条件即可
GET z/_search
{
"query": {
"bool": {
"should": [
{
"term": {
"address": "湖北湖南"
}
},
{
"range": {
"age": {
"gte": 30,
"lte": 36
}
}
}
]
}
}
}
8.9：高亮查询【重要】
单个域的高亮
GET z/_search
{
"query": {
"match": {
"address": "湖北教育"
}
},
"highlight": {
"fields": {
"address": {}
},
"pre_tags": "<font color='red'>",
"post_tags": "</font>"
}
}
多个域的高亮
GET z/_search
{
"query": {
"bool": {
"should": [
{
"match": {
"address": "湖北"
}
},
{
"term": {
"job": "工程师"
}
}
]
}
},
"highlight": {
"fields": {
"address": {},
"email": {}
},
"pre_tags": "<font color='red'>",
"post_tags": "</font>"
}
}
8.10：boosting查询
影响文档分数的因素:
1.当查询的关键字在文档出现的频次越高，分数越高
2.指定的文档内容越短，分数越高,如查找的是高富帅，指定文档内容就是高富帅
GET es_user/_search
{
"query": {
"boosting": {
"positive": {
"match": {
"address": "湖北省"
}
},
"negative": {
"match": {
"job": "前端"
}
},
"negative_boost": 0.001
}
}
}
8.11：过滤【重要】
查询结果过滤掉不想要的字段
select *
select age
GET /es_user/_search
{
"_source": ["age","gender"],
"query": {
"match_all": {}
}
}
查询结果过滤掉不想要的文档
过滤和查询都能起到对结果集的过滤效果，但是查询会影响到文档的评分及排名，而过滤不会。
如果我们需要在查询结果中进行过滤，并且不希望过滤条件影响评分，那么就不要把过滤条件作为查询条件来用。而是使用
filter方式
复合查询影响分数演示
需求：address中有湖北词条 must job中有工程师词条
复合查询
GET es_user/_search
{
"query": {
"bool": {
"must": [
{
"match": {
"address": "湖北"
}
},
{
"term": {
"job.keyword": "开发工程师"
}
}
]
}
}
}
过滤(不会改变分数)
GET es_user/_search
{
"query": {
"bool": {
"must": [
{
"match": {
"address": "湖北"
}
}
],
"filter": [
{
"term": {
"job.keyword": "开发工程师"
}
}
]
}
}
}
8.12：排序【重要】
GET /es_user/_search
{
"query": {
"bool": {
"must": [
{
"match": {
"address": "武汉人"
}
}
],
"filter": [
{
"term": {
"job": "工程师"
}
}
]
}
},
"sort": [
{
"age": {
"order": "desc"
}
}
]
}
聚合【重要】
Elasticsearch中的聚合，包含多种类型，最常用的两种，一个叫桶，一个叫度量：
度量（metrics）
分组完成以后，我们一般会对组中的数据进行聚合运算，例如求平均值、最大、最小、求和等，这些在ES中称为度量
比较常用的一些度量聚合方式：
Avg Aggregation：求平均值
Max Aggregation：求最大值
Min Aggregation：求最小值
Percentiles Aggregation：求百分比
Stats Aggregation：同时返回avg、max、min、sum、count等
Sum Aggregation：求和
Top hits Aggregation：求前几
Value Count Aggregation：求总数
……
GET /es_user/_search
{
"query": {
"match": {
"address": "武汉"
}
},
"aggs": {
"age_avg": {
"avg": {
"field": "age"
}
},
"max_age":{
"max": {
"field": "age"
}
}
}
}
桶（bucket）
桶的作用，是按照某种方式对数据进行分组，每一组数据在ES中称为一个桶 比如：职位
GET /es_user/_search
{
"query": {
"match": {
"address": "武汉市"
}
},
"aggs": {
"job_group": {
"terms": {
"field": "job.keyword",
"size": 10
}
}
}
}
pom依赖
<dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-data-elasticsearch</artifactId> </dependency>
建立索引库
@SpringBootTest
@RunWith(SpringRunner.class)
public class ESAppTest {
@Autowired
private ElasticsearchRestTemplate template;
//创建索引
@Test
public void testIndex (){
//创建索引
template.createIndex(ESUser.class);
//mapping设置（设置域的类型）
template.putMapping(ESUser.class);
}
}
9.3.2：保存文档
//保存文档
@Test
public void testSave (){
ESUser doc = new ESUser();
doc.setId(1);
doc.setAddress("武汉千锋");
doc.setCity("武汉市");
doc.setAge(12);
ESUser save = template.save(doc);
}
9.3.3：刪除文档
//删除文档
@Test
public void testDelete (){
template.delete("1",ESUser.class);
}
9.3.4：查询（match_all）
//match_all
@Test
public void test3(){
MatchAllQueryBuilder matchAllQueryBuilder = QueryBuilders.matchAllQuery();
NativeSearchQuery build = new NativeSearchQueryBuilder()
//指定查询方式
.withQuery(matchAllQueryBuilder)
.build();
SearchHits<ESUser> search = template.search(build, ESUser.class);
//获取总记录数
long totalHits = search.getTotalHits();
System.out.println(totalHits);
//获取文档列表
List<SearchHit<ESUser>> searchHits = search.getSearchHits();
List<ESUser> list =
searchHits.stream().map(hit->hit.getContent()).collect(Collectors.toList());
list.forEach(System.out::println);
}
9.3.5：查询（match）
//match
@Test
public void test4(){
MatchQueryBuilder matchQueryBuilder = QueryBuilders.matchQuery("address", "我喜欢湖北");
NativeSearchQuery build = new NativeSearchQueryBuilder()
//指定查询方式
.withQuery(matchQueryBuilder)
.build();
SearchHits<ESUser> search = template.search(build, ESUser.class);
//获取总记录数
long totalHits = search.getTotalHits();
System.out.println(totalHits);
//获取文档列表
List<SearchHit<ESUser>> searchHits = search.getSearchHits();
List<ESUser> list =
searchHits.stream().map(hit->hit.getContent()).collect(Collectors.toList());
list.forEach(System.out::println);
}
9.3.6：查询（term）
//term
@Test
public void test5(){
TermQueryBuilder termQueryBuilder = QueryBuilders.termQuery("address", "湖北");
NativeSearchQuery build = new NativeSearchQueryBuilder()
//指定查询方式
.withQuery(termQueryBuilder)
.build();
SearchHits<ESUser> search = template.search(build, ESUser.class);
//获取总记录数
long totalHits = search.getTotalHits();
System.out.println(totalHits);
//获取文档列表
List<SearchHit<ESUser>> searchHits = search.getSearchHits();
List<ESUser> list =
searchHits.stream().map(hit->hit.getContent()).collect(Collectors.toList());
list.forEach(System.out::println);
}
9.3.7：查询(range)
//range
@Test
public void test6(){
RangeQueryBuilder age = QueryBuilders.rangeQuery("age").lte(36).gte(30);
NativeSearchQuery build = new NativeSearchQueryBuilder()
//指定查询方式
.withQuery(age)
.build();
SearchHits<ESUser> search = template.search(build, ESUser.class);
//获取总记录数
long totalHits = search.getTotalHits();
System.out.println(totalHits);
//获取文档列表
List<SearchHit<ESUser>> searchHits = search.getSearchHits();
List<ESUser> list =
searchHits.stream().map(hit->hit.getContent()).collect(Collectors.toList());
list.forEach(System.out::println);
}
9.3.8：查询（page）
//page
@Test
public void test7(){
RangeQueryBuilder age = QueryBuilders.rangeQuery("age").lte(36).gte(0);
PageRequest pageRequest = PageRequest.of(0, 2);
NativeSearchQuery build = new NativeSearchQueryBuilder()
//指定查询方式
.withQuery(age)
//设置分页信息
.withPageable(pageRequest)
.build();
SearchHits<ESUser> search = template.search(build, ESUser.class);
//获取总记录数
long totalHits = search.getTotalHits();
System.out.println(totalHits);
//获取文档列表
List<SearchHit<ESUser>> searchHits = search.getSearchHits();
List<ESUser> list =
searchHits.stream().map(hit->hit.getContent()).collect(Collectors.toList());
list.forEach(System.out::println);
}
9.3.9：查询(bool)
/**
* address 我喜欢湖北 match
* should
* 30>age>10 range
*/
@Test
public void test8(){
BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
MatchQueryBuilder matchQueryBuilder = QueryBuilders.matchQuery("address", "我喜欢湖北");
RangeQueryBuilder age = QueryBuilders.rangeQuery("age").gte(10).lte(30);
boolQueryBuilder.should(matchQueryBuilder).should(age);
NativeSearchQuery build = new NativeSearchQueryBuilder()
//指定查询方式
.withQuery(boolQueryBuilder)
.build();
SearchHits<ESUser> search = template.search(build, ESUser.class);
//获取总记录数
long totalHits = search.getTotalHits();
System.out.println(totalHits);
//获取文档列表
List<SearchHit<ESUser>> searchHits = search.getSearchHits();
List<ESUser> list =
searchHits.stream().map(hit->hit.getContent()).collect(Collectors.toList());
list.forEach(System.out::println);
}
9.3.10：查询(high light)
//单域highlight
@Test
public void test9(){
MatchQueryBuilder matchQueryBuilder = QueryBuilders.matchQuery("address", "我喜欢湖北");
NativeSearchQuery build = new NativeSearchQueryBuilder()
//指定查询方式
.withQuery(matchQueryBuilder)
//设置高亮
.withHighlightBuilder(getHighlightBuilder("address"))
.build();
SearchHits<ESUser> search = template.search(build, ESUser.class);
//获取总记录数
long totalHits = search.getTotalHits();
System.out.println(totalHits);
//获取文档列表
List<SearchHit<ESUser>> searchHits = search.getSearchHits();
List<ESUser> list =
// searchHits.stream().map(hit->hit.getContent()).collect(Collectors.toList());
searchHits.stream().map(hit->{
ESUser esUser = hit.getContent();
//取高亮
Map<String, List<String>> highlightFields = hit.getHighlightFields();
highlightFields.forEach((k,v)->{
String highlightVal = v.get(0);
if("address".equals(k)){
esUser.setAddress(highlightVal);
}
});
return esUser;
}).collect(Collectors.toList());
list.forEach(System.out::println);
}
//多域highlight
@Test
public void test10(){
MatchQueryBuilder addressmatchQueryBuilder = QueryBuilders.matchQuery("address", "我喜欢湖北");
MatchQueryBuilder jobmatchQueryBuilder = QueryBuilders.matchQuery("job", "工程师");
BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
boolQueryBuilder.should(addressmatchQueryBuilder).should(jobmatchQueryBuilder);
NativeSearchQuery build = new NativeSearchQueryBuilder()
//指定查询方式
.withQuery(boolQueryBuilder)
//设置高亮
.withHighlightBuilder(getHighlightBuilder("address","job"))
.build();
SearchHits<ESUser> search = template.search(build, ESUser.class);
//获取总记录数
long totalHits = search.getTotalHits();
System.out.println(totalHits);
//获取文档列表
List<SearchHit<ESUser>> searchHits = search.getSearchHits();
List<ESUser> list =
// searchHits.stream().map(hit->hit.getContent()).collect(Collectors.toList());
searchHits.stream().map(hit->{
ESUser esUser = hit.getContent();
//取高亮
Map<String, List<String>> highlightFields = hit.getHighlightFields();
highlightFields.forEach((k,v)->{
String highlightVal = v.get(0);
if("address".equals(k)){
esUser.setAddress(highlightVal);
}
if("job".equals(k)){
esUser.setJob(highlightVal);
}
});
return esUser;
}).collect(Collectors.toList());
list.forEach(System.out::println);
}
// 设置高亮字段
private HighlightBuilder getHighlightBuilder(String... fields) {
// 高亮条件
HighlightBuilder highlightBuilder = new HighlightBuilder(); //生成高亮查询器
for (String field : fields) {
highlightBuilder.field(field);//高亮查询字段
}
highlightBuilder.requireFieldMatch(false); //如果要多个字段高亮,这项要为false
highlightBuilder.preTags("<span style=\"color:red\">"); //高亮设置
highlightBuilder.postTags("</span>");
//下面这两项,如果你要高亮如文字内容等有很多字的字段,必须配置,不然会导致高亮不全,文章内容缺失等
highlightBuilder.fragmentSize(800000); //最大高亮分片数
highlightBuilder.numOfFragments(0); //从第一个分片获取高亮片段
return highlightBuilder;
}
GET es_user_3/_search
{
"query": {
"bool": {
"should": [
{
"match": {
"address": "我喜欢湖北"
}
},
{
"match": {
"job": "工程师"
}
}
]
}
},
"highlight": {
"fields": {
"address": {},
"job": {}
},
"pre_tags": "<font style='color:red'>",
"post_tags": "</font>"
}
}