elasticsearch 介绍、查询及使用（must、should）
作者：互联网
创建索引
{
"mappings": {
"properties": {
"id": {"type":"keyword"},
"pname": {"type":"text", "analyzer": "ik_max_word", "search_analyzer": "ik_max_word","fields": {"keyword":{"type":"keyword"}}},
"title": {"type":"text", "analyzer": "ik_max_word", "search_analyzer": "ik_max_word","fields": {"keyword":{"type":"keyword"}}},
"describe": {"type":"text", "analyzer": "ik_max_word", "search_analyzer": "ik_max_word","fields": {"keyword":{"type":"keyword"}}},
"rich_text": {"type":"text", "analyzer": "ik_max_word", "search_analyzer": "ik_max_word","fields": {"keyword":{"type":"keyword"}}},
"update_time": { "type": "date", "format": "yyyy-MM-dd HH:mm:ss||yyyy-MM-dd||epoch_millis"}
}
}
}
@Id
@Field(index=true, store = true, type = FieldType.Keyword)
private String id;
/**
* 更新时间
*/
@JSONField(serialize = false)
@Field(index = true, store = true, type = FieldType.Date,format = DateFormat.custom,pattern = "yyyy-MM-dd HH:mm:ss")
private Date update_time;
/**
* 标题
*/
@Field(index = true, analyzer = "ik_max_word", store = true, searchAnalyzer = "ik_smart", type = FieldType.Text)
private String title;
/** 名称*/
@Field(index = true, analyzer = "ik_max_word", store = true, searchAnalyzer = "ik_smart", type = FieldType.Text)
private String pname;
/**
* 摘要
*/
@Field(index = true, analyzer = "ik_max_word", store = true, searchAnalyzer = "ik_smart", type = FieldType.Text)
private java.lang.String describe;
/**
* 文章正文
*/
@JSONField(serialize = false)
@Field(index = true, analyzer = "ik_max_word", store = true, searchAnalyzer = "ik_smart", type = FieldType.Text)
private String rich_text;
查询介绍
must 返回的文档必须满足must子句的条件,并且参与计算分值
filter 返回的文档必须满足filter子句的条件,但是不会像must一样,参与计算分值
should 返回的文档可能满足should子句的条件.在一个bool查询中,如果没有must或者filter,有一个或者多个should子句,那么只要满足一个就可以返回.
must_not 返回的文档必须不满足定义的条件
matchAllQuery匹配所有
termQuery精准匹配，大小写敏感且不支持
matchPhraseQuery对中文精确匹配
matchQuery("key", Obj) 单个匹配, field不支持通配符, 前缀具高级特性
multiMatchQuery("text", "field1", "field2"..); 匹配多个字段, field有通配符忒行
实现匹配模糊查询和等值查询
模糊查询
BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
boolQueryBuilder.should(QueryBuilders.wildcardQuery("title", "*" + keyword + "*"));
等值查询 必须保存此字段有 “字段”.keyword 索引
BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
boolQueryBuilder.should(QueryBuilders.matchQuery("title.keyword", keyword))
注意：当使用should查询时，如果包含了must或者filter查询，那么should的查询语句就不是或者的意思了,而是有或者没有都行的含义。
例如：在a=1且b=2的数据中，找出c=1或者d=2的数据
{
"query": {
"bool": {
"must": [{
"match": {
"a": "1",
"b": "2"
}
}],
"should": [{
"match": {
"c": "1"
}
},
{
"match": {
"d": "2"
}
}
]
}
}
}
这样写的时候should是没有用的，这是新手可能犯的错误之一。 在编写查询条件的时候，不能用口头上的逻辑进行编写，而是要换成数学逻辑才能进行执行（数据库同理）。 如上例，数学逻辑应该是 （a==1&&b==2&&c==1）||(a==1&&b==2&&d==2)（java and c语言版），这样的结构去查询。
实现方式一：
QueryBuilder query = QueryBuilders.boolQuery()
.should(QueryBuilders.boolQuery()
.filter(QueryBuilders.termQuery("a", 1))
.filter(QueryBuilders.termQuery("b", 2))
.filter(QueryBuilders.termQuery("c", 1))
.should(QueryBuilders.boolQuery()
.filter(QueryBuilders.termQuery("a", 1))
.filter(QueryBuilders.termQuery("b", 2))
.filter(QueryBuilders.termQuery("d", 2));
{
"query": {
"bool": {
"should": [{
"match": {
"a": "1",
"b": "2",
"c": "1"
}
}],
"should": [{
"match": {
"a": "1",
"b": "2",
"d": "2"
}
}]
}
}
}
实现方式二：
BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
boolQueryBuilder.should(QueryBuilders.matchQuery("title", keyword))
boolQueryBuilder.should(QueryBuilders.matchQuery("pname", keyword));
//要求至少满足一个Should条件
boolQueryBuilder.minimumShouldMatch(1);
{
"bool" : {
"must" : [
{
"term" : {
"site_id" : {
"value" : 110,
"boost" : 1.0
}
}
}
],
"should" : [
{
"match" : {
"title" : {
"query" : "咖啡",
"operator" : "OR"
}
}
},
{
"match" : {
"pname" : {
"query" : "咖啡",
"operator" : "OR"
}
}
},
],
"minimum_should_match" : "1"
}
}
实现方式三：
String[] searchArray = new String[]{"title","pname"};
searchKeys.toArray(searchArray);
boolQueryBuilder.must(QueryBuilders.multiMatchQuery(keyword, searchArray).operator(Operator.AND));
{
"bool" : {
"must" : [
{
"term" : {
"site_id" : {
"value" : 110,
"boost" : 1.0
}
}
},
{
"multi_match" : {
"query" : "咖啡",
"fields" : [
"pname^1.0",
"title^1.0"
],
"operator" : "AND"
}
}
]
}
}
标签：keyword,analyzer,should,ik,elasticsearch,type,QueryBuilders,must 来源： https://www.cnblogs.com/shanheyongmu/p/16470371.html