elasticsearch入门使用(三) Query DSL

Elasticsearch Reference [6.2] » Query DSL
参考官方文档 :https://www.elastic.co/guide/en/elasticsearch/reference/current/query-filter-context.html

一、组合查询 Compound queries

  1. Constant Score Query 指定_score分数查询
GET /_search
{
    "query": {
        "constant_score" : {
            "filter" : {
                "term" : { "user" : "kimchy"}
            },
            "boost" : 1.2
        }
    }
}
  1. Bool Query 布尔值查询

must:查询的条件必须在匹配的文档中,并计算相似度得分
filter:必须满足条件,不会计算相似度得分
should:满足子条件的一个或者多个,满足的格式可以通过"minimum_should_match" : 1设置,类似 OR (如果查询中包filter则至少满足一个should)
must_not:返回的文档必须不满足条件,类似 NOT

tips : 日期格式在添加文档和搜索的时候加上T,字符串不区分大小写 如pek

GET /stu/_search
{
   "query": { 
    "bool": { 
      "must": [
        { "match": { "address":"上海市 保德路 闸北区"}}
      ],
      "filter": [
        {"term":{ "id": 11 }},
        {"term":{ "city": "pek" }},
        {"range":{"regdate": {"gte": "2018-03-03T15:33:32","lte":"2018-03-03T15:33:33"}}} 
      ],
      "should":[
        {"term":{ "score": 80.8 }},
        {"term":{ "score": 80.0 }}
      ],
      "must_not":[{
        "term" : { "age" : 30 }
      }],
      "minimum_should_match" : 1,
      "boost" : 1.0
    }
  }
}

二、查询上下文

Query Context:文档和查询条件的匹配度,出了决定是否与文档匹配外,还会计算查询条件和文档的匹配度_score

GET /_search
{
    "query" : {
        "term" : { "user" : "kimchy" }
    }
}

三、过滤上下文

Filter context: 精确搜索文档和查询是否匹配,不会去计算匹配度,主要用于过滤结构化数据.
经常使用的过滤器会被elasticsearch自动缓存,以提高查询效率

示例:title中包含search、content中包含elasticsearch 且 status="published" & publish_date >="2015-01-01"
filter里的term、range

GET /_search
{
  "query": { 
    "bool": { 
      "must": [
        { "match": { "title":   "Search"}}, 
        { "match": { "content": "Elasticsearch" }}  
      ],
      "filter": [ 
        { "term":  { "status": "published" }}, 
        { "range": { "publish_date": { "gte": "2015-01-01" }}} 
      ]
    }
  }
}

四、查询所有

查询所有_score的文档,(注:boost的默认值是 1.0)

GET /_search
{
    "query": {
        "match_all": {}
    }
}

查询_score=1.2的所有文档

GET /_search
{
    "query": {
        "match_all": { "boost" : 1.2 }
    }
}

五、全文查询

1.match query

query:查询字段message中包含 " this is test " , 注意 "to be or not to be" 属于停顿词,过滤器默认会把remove调,设置zero_terms_query:"all"启用
另operator/zero_terms_query非必填参数,更详细内容查看match query官方文档

GET /_search
{
    "query": {
        "match" : {
            "message" : {
                "query" : "to be or not to be,this is test",
                "operator" : "and",
                "zero_terms_query": "all"
            }
        }
    }
}

2.Match Phrase Query

match_phrase从分析文本"this is a test"中创建一组词去查询,analyzer分词器也可以使用ik等中文分词器

GET /_search
{
    "query": {
        "match_phrase" : {
            "message" : {
                "query" : "this is a test",
                "analyzer" : "standard"
            }
        }
    }
}

3.Match Phrase Prefix Query

match_phrase_prefix与match_phrase类似,只是它允许在文本中的最后一个词的前缀匹配

GET /_search
{
    "query": {
        "match_phrase_prefix" : {
            "message" : "quick brown f"
        }
    }
}

4.Multi Match Query

multi_match匹配查询上以允许多字段查询(subject/message字段):

GET /_search
{
  "query": {
    "multi_match" : {
      "query":    "this is a test", 
      "fields": [ "subject", "message" ] 
    }
  }
}

5.Common Terms Query
common:停顿词配置相关如 " the to be "等

6.Query String Query
query_string:没理解看官方文档

7.Simple Query String Query
simple_query_string: 与query_string查询不同的是,simple_query_string查询永远不会抛出异常,并放弃查询的无效部分

GET /_search
{
  "query": {
    "simple_query_string" : {
        "query": "\"fried eggs\" +(eggplant | potato) -frittata",
        "fields": ["title^5", "body"],
        "default_operator": "and"
    }
  }
}
posted @ 2018-03-13 16:30  nickchou  阅读(829)  评论(0编辑  收藏  举报