ES 7.9.0 初探之常见操作-match
文章目录
-
- 1.前缀查询（prefix）
- 2. 短语匹配（match_phrase）
- 3.短语前缀匹配查询(match_phrase_prefix)
- 4.多重匹配查询（Multi-match query）
- 5.是否匹配前缀（match_bool_prefix）
上一篇我们使用ES进行了简单的CURD操作，今天来进行一点简单的匹配查询的操作。
首先来看一下我编了哪些数据：
GET yytest/_search
得到的数据有
{
"took" : 0,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 5,
"relation" : "eq"
},
"max_score" : 1.0,
"hits" : [
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "l5PudHUBTpZvBmMg9zdF",
"_score" : 1.0,
"_source" : {
"name" : "zhangsan",
"age" : 23,
"birthday" : "1998-08-28"
}
},
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "kZU9jnUBTpZvBmMggwYh",
"_score" : 1.0,
"_source" : {
"name" : "jerry David",
"age" : 24,
"birthday" : "1996-03-12",
"description" : "he likes playing LOL"
}
},
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "tJU-jnUBTpZvBmMg6gZo",
"_score" : 1.0,
"_source" : {
"name" : "lucy Allen",
"age" : 23,
"birthday" : "1997-09-22",
"description" : "she is pretty"
}
},
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "dpU8jnUBTpZvBmMgfAZR",
"_score" : 1.0,
"_source" : {
"name" : "tom green",
"age" : "23",
"birthday" : "1995-08-18",
"description" : "he is tall and handsome, and tony is his uncle"
}
},
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "LJZNonUBTpZvBmMgSoqi",
"_score" : 1.0,
"_source" : {
"name" : "tony stack",
"age" : "34",
"birthday" : "1986-09-21",
"description" : "he likes playing basketball, his last name is stack"
}
}
]
}
}
1.前缀查询（prefix）
查询语句
GET yytest/_search
{
"query":{
"prefix": {
"description": "he"
}
}
}
这样的GET请求将会得到下面的返回结果
{
"took" : 1,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 3,
"relation" : "eq"
},
"max_score" : 1.0,
"hits" : [
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "kZU9jnUBTpZvBmMggwYh",
"_score" : 1.0,
"_source" : {
"name" : "jerry David",
"age" : 24,
"birthday" : "1996-03-12",
"description" : "he likes playing LOL"
}
},
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "dpU8jnUBTpZvBmMgfAZR",
"_score" : 1.0,
"_source" : {
"name" : "tom green",
"age" : "23",
"birthday" : "1995-08-18",
"description" : "he is tall and handsome, and tony is his uncle"
}
},
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "LJZNonUBTpZvBmMgSoqi",
"_score" : 1.0,
"_source" : {
"name" : "tony stack",
"age" : "34",
"birthday" : "1986-09-21",
"description" : "he likes playing basketball, his last name is stack"
}
}
]
}
}
从返回的数据中，我们可以看到，hit了3条数据。
注意：
如果的ES版本比较旧的，即7.0以前的，查询语句需要指定type，但是7.0之后的版本已经逐步废弃这一用法了。如果用以下的查询语句，即
GET yytest/_doc/_search
{
"query":{
"prefix": {
"description": "he"
}
}
}
也能够返回数据，但是最上面会显示如下的内容
#! Deprecation: [types removal] Specifying types in search requests is deprecated.
因此我们不要再指定type了。
2. 短语匹配（match_phrase）
查询description中的短语包括“likes playing”
查询的语句是
GET yytest/_search
{
"query": {
"match_phrase": {
"description": "likes playing"
}
}
}
返回的结果是
{
"took" : 1,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 2,
"relation" : "eq"
},
"max_score" : 2.0482664,
"hits" : [
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "kZU9jnUBTpZvBmMggwYh",
"_score" : 2.0482664,
"_source" : {
"name" : "jerry David",
"age" : 24,
"birthday" : "1996-03-12",
"description" : "he likes playing LOL"
}
},
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "LJZNonUBTpZvBmMgSoqi",
"_score" : 1.4778953,
"_source" : {
"name" : "tony stack",
"age" : "34",
"birthday" : "1986-09-21",
"description" : "he likes playing basketball, his last name is stack"
}
}
]
}
}
3.短语前缀匹配查询(match_phrase_prefix)
短语前缀匹配查询，即经过分词器分词之后，最后一个词作为搜素的前缀。举例来说
GET yytest/_search
{
"query": {
"match_phrase_prefix": {
"description": {
"max_expansions": 10,
"query": "he likes play"
}
}
}
}
这里的就是搜索短语，短语的前两个单词是“he likes ”，然后以第三个单词play作为前缀条件，继续搜索。返回的结果
{
"took" : 1,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 2,
"relation" : "eq"
},
"max_score" : 2.3848,
"hits" : [
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "kZU9jnUBTpZvBmMggwYh",
"_score" : 2.3848,
"_source" : {
"name" : "jerry David",
"age" : 24,
"birthday" : "1996-03-12",
"description" : "he likes playing LOL"
}
},
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "LJZNonUBTpZvBmMgSoqi",
"_score" : 1.720716,
"_source" : {
"name" : "tony stack",
"age" : "34",
"birthday" : "1986-09-21",
"description" : "he likes playing basketball, his last name is stack"
}
}
]
}
}
可以看到，匹配了两条数据。
然后“max_expansions"用来限制匹配道的文档的数量，默认值为50。即如果素以库中有20条数据，限制了max_expansions为10，则会按存储顺序只返回前10条数据。不过这里我试了一下，貌似有点问题，当我将其限制为1条的时候，还是会查处两条数据，即我当前索引库的所有数据。
max_expansions讲解了有可能的原因，但是我测试的结果不符合预期，有懂的大佬可以指点我一下。
4.多重匹配查询（Multi-match query）
多重匹配查询是匹配查询多个字段，例如查询name和description字段中包含tony的文档，即
GET yytest/_search
{
"query": {
"multi_match": {
"query": "tony",
"fields": ["name","description"]
}
}
}
返回的结果是
{
"took" : 0,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 2,
"relation" : "eq"
},
"max_score" : 1.567127,
"hits" : [
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "LJZNonUBTpZvBmMgSoqi",
"_score" : 1.567127,
"_source" : {
"name" : "tony stack",
"age" : "34",
"birthday" : "1986-09-21",
"description" : "he likes playing basketball, his last name is stack"
}
},
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "dpU8jnUBTpZvBmMgfAZR",
"_score" : 1.1083853,
"_source" : {
"name" : "tom green",
"age" : "23",
"birthday" : "1995-08-18",
"description" : "he is tall and handsome, and tony is his uncle"
}
}
]
}
}
可以看到，这里包含了两个文档，其中是那么字段中包含tony的文档LJZNonUBTpZvBmMgSoqi，和描述中国包含tony的文档dpU8jnUBTpZvBmMgfAZR。
5.是否匹配前缀（match_bool_prefix）
match_bool_prefix与match_phrase_prefix的区别在于match_bool_prefix分词后，不管单词的前后顺序，即如果匹配“he likes”，那么“likes he”，“he“和"likes"都会匹配到。
GET yytest/_search
{
"query":{
"match_bool_prefix":{
"description":"tony stack"
}
}
}
返回的结果是
{
"took" : 1,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 2,
"relation" : "eq"
},
"max_score" : 1.1083853,
"hits" : [
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "dpU8jnUBTpZvBmMgfAZR",
"_score" : 1.1083853,
"_source" : {
"name" : "tom green",
"age" : "23",
"birthday" : "1995-08-18",
"description" : "he is tall and handsome, and tony is his uncle"
}
},
{
"_index" : "yytest",
"_type" : "_doc",
"_id" : "LJZNonUBTpZvBmMgSoqi",
"_score" : 1.0,
"_source" : {
"name" : "tony stack",
"age" : "34",
"birthday" : "1986-09-21",
"description" : "he likes playing basketball, his last name is stack"
}
}
]
}
}
最后给出官网的地址ES 匹配，大家可以自行查阅。