IK分词器的分词

ik_max_word

有IK分词器提供，分化的程度按照插件的自行判断，分得更加细而已。
缺点：过于分化，多余条件检索

#方式一ik_max_word 粗密度
GET /_analyze
{
  "analyzer": "ik_max_word",
  "text": "LOL全球总决赛No.1"
}

ik_smart

少量关键字，推荐使用这个。

 #方式二ik_smart 细密度
GET /_analyze
{
  "analyzer": "ik_smart",
  "text": "LOL全球总决赛No.1"
}

准备测试数据

# 添加索引
PUT person2
{
  "mappings": {
    "properties": {
      "name": {
        "type": "keyword"
      },
      "address": {
        "type": "text",
        "analyzer": "ik_max_word"
      }
    }
  }
}
# 添加文档
POST /person2/_doc/1
{
  "name":"张三",
  "age":18,
  "address":"重庆市江北区"
}
POST /person2/_doc/2
{
  "name":"李四",
  "age":19,
  "address":"重庆市渝中区"
}
POST /person2/_doc/3
{
  "name":"王五",
  "age":20,
  "address":"重庆市南岸区"
}
POST /person2/_doc/4
{
  "name":"何六",
  "age":21,
  "address":"四川省高新区"
}
POST /person2/_doc/5
{
  "name":"田七",
  "age":22,
  "address":"四川省双流区"
}
POST /person2/_doc/7
{
  "name":"黄黄",
  "age":25,
  "address":"上海市长宁区"
}

单个词条检索 [ term ]

# 单个词条检索
# ik分词器分后的词条进行检索
# keyword类型必须完全一致，才能查询出
# text类型会进行单个相匹配的词条查询
GET /person2/_search
{
  "query": {
    "term": {
      "address": {
        "value": "南岸"
      }
    }
  }
}

多词条检索 [ terms ]

 # 多词条检索
GET /person2/_search
{
  "query": {
    "terms": {
      "address": ["重庆","区"]
    }
  }
}

模糊查询 [ fuzzy ]

# 模糊查询
# 允许关键字出现两个字符的偏差，默认偏差为1
# fuzziness：设置偏差
GET /person2/_search
{
  "query": {
    "fuzzy": {
      "address": {
        "value": "南岸12",
        "fuzziness":"2"
      }
    }
  }
}

全文查询 [ match ]

# 条件检索 [重庆][南岸]
# 对关键字进行分词
# 默认取并集（or）
GET /person2/_search
{
  "query": {
    "match": {
      "address":"重庆南岸"
    }
  }
}
# 交集（and）
GET /person2/_search
{
  "query": {
    "match": {
      "address": {
        "query": "重庆南岸",
        "operator": "and"
      }
    }
  }
}

并集（or）

交集（and）

全部（ match_all）

# 全文检索
GET /person2/_search
{
    "query": { "match_all": {} }
}

多字段检索 [ multi_match ]

# 多字段检索 [address] [name] 只要含有“重庆”两字关键字，查询出相应文档
GET /person2/_search
{
  "query": {
    "multi_match": {
      "query": "重庆",
      "fields": ["address","name"]
    }
  }
}

通配符检索 [ wildcard ]

# 模糊查询
# "重*"以“重”开头所有关键字
GET /person2/_search
{
  "query": {
    "wildcard": {
      "address":"重*"
    }
  }
}

# 模糊查询
# "岸?"以“岸”开头所有关键字
GET /person2/_search
{
  "query": {
    "wildcard": {
      "address":"岸?"
    }
  }
}

正则查询 [ regexp ]

正则查询取决于正则表达式的效率

# 正则查询
# (.)*为任意字符
GET /person2/_search
{
  "query": {
    "regexp": {
      "address":"(.)*"
    }
  }
}

前缀查询 [ prefix ]

# 前缀查询
# 末尾字查询结果为无
GET /person2/_search
{
  "query": {
    "prefix": {
      "address":"上"
    }
  }
}

范围查询 [ range ]

# 范围查询
# gt大于 gte大于等于 lt小于 lte小于等于
GET /person2/_search
{
  "query": {
    "range": {
      "age":{
        "gte":18,
        "lte":20
      }
    }
  }
}

多条件查询 [ query_string ]

注意：看着下面两个查询，除了“重庆”和“上海”关键字不同外，再无区别。而添加的文档里面，两种类型的数据都是存在，并且分词器，也成功将“重庆市”和“上海市”分词，结果却大出意外。

# 多条件查询
# 识别单个词条 and 两个词条都要查询，得到交集
# or 查询出并集
GET /person2/_search
{
  "query": {
    "query_string": {
      "default_field": "address",
      "query": "重庆 AND 市 OR 区"
    }
  }
}
GET /person2/_search
{
  "query": {
    "query_string": {
      "default_field": "address",
      "query": "上海 AND 市 OR 区"
    }
  }
}

上海的数据没有查询出来！！！

结论：query_string查询数据有遗漏问题，选择使用下面的方式补救。

自动分词的连接符 [ default_operator ]

# 自动分词的连接符
GET /person2/_search
{
  "query": {
    "query_string": {
      "default_field": "address", 
      "default_operator": "AND",
      "query": "上海市"
    }
  }
}

多条件多词条查询 （or交集）

GET /person2/_search
{
  "query": {
    "query_string": {
      "fields": ["address"],
      "query": "上海 OR 市"
    }
  }
}

单例多条件查询 [ simple_query_string ]

# 不识别连接符（and、or），分别将[重庆][市][区]词条以and查询
GET /person2/_search
{
  "query": {
    "simple_query_string": {
      "query": "重庆 AND 市 OR 区",
      "fields": ["address"]
    }
  }
}

布尔查询 [ bool ]

must

# 布尔查询
# must：条件必须成立
# must_not：条件必须不成立
# _score（得分）：每条数据匹配度，匹配度越高，得分越高
GET /person2/_search
{
  "query": {
    "bool": {
      "must": [
        {
          "term": {
            "address": {
              "value": "重庆"
            }
          }
        }
      ]
    }
  }
}

filter

# filter：过滤
# _score（得分）不显示，性能更优
GET /person2/_search
{
  "query": {
    "bool": {
      "filter": [
        {
          "term": {
            "address": {
              "value": "重庆"
            }
          }
        }
      ]
    }
  }
}

should

# should：或许
GET /person2/_search
{
  "query": {
    "bool": {
      "should": [
        {
          "terms": {
            "address": [
              "河南",
              "河北",
              "重庆"
            ]
          }
        }
      ]
    }
  }
}

组合使用

# 组合
GET /person2/_search
{
  "query": {
    "bool": {
      "must": [
        {
          "term": {
            "address": {
              "value": "重庆"
            }
          }
        }
      ],
      "filter": [
        {
          "term": {
            "address": {
              "value": "南岸"
            }
          }
        }
      ]
    }
  }
}

倒序 [ sort ]

# id 倒序 asc / desc
GET /person2/_search
{
  "query": { "match_all": {} },
  "sort": [
    { "_id": "desc" }
  ]
}

关键字查询

# 关键字查询
# 查询name和张三关键字，
# 张三不存在，默认忽视
GET /person2/_search
{
  "query": { "match_all": {} },
  "_source": ["name","张三"]
}

排除关键字查询

# 排除关键字查询
GET /person2/_search
{
  "_source": {
    "excludes":"age"
  },
  "query": { "match_all": {} }
}

分页查询

# 分页查询 
# from：当前页，size：总页数
GET /person2/_search
{
  "query": { "match_all": {} },
  "from": 3,
  "size": 3,
  "sort": {  "age": "asc"  }
}

桶聚合 [ aggs ]

# 桶聚合  分组
# 格式
#  "aggs": {
#    "自定义属性名": {
#      "terms": {
#        "field": "属性",
#        "size": 10
#      }
#    }
#  }
GET person2/_search
{
  "size": 0,
  "query": {
    "match": {
      "address": "重庆"
    }
  },
  "aggs": {
    "address_name": {
      "terms": {
        "field": "name",
        "size": 10
      }
    }
  }
}

指标聚合 [ aggs ]

max,avg,min,sun….

# 指标聚合 聚合函数
# 四川省内最大年纪的人
GET person2/_search
{
  "query": {
    "match": {
      "address": "四川"
    }
  },
  "aggs": {
    "max_age": {
      "max": {
        "field": "age"
      }
    }
  }
}

嵌套查询

GET person2/_search
{
  "size": 0, 
  "aggs": {
    "my_name": {
      "terms": {
        "field": "name"
      },
      "aggs": {
        "age_max":{
          "max": {
            "field": "age"
          }
        }
      }
    } 
  }
}

# 嵌套不能查询keyword类型
GET person2/_search
{
  "size": 0, 
  "aggs": {
    "age_max":{
      "max": {
        "field": "age"
      }
    },
    "my_name":{
      "terms": {
        "field": "name"
      }
    }
  }
}

阶梯分桶 [ histogram ]

# 阶梯分桶
# interval：阶梯值
# min_doc_count：最小文档数
GET /person2/_search
{
  "size": 0,
  "aggs": {
    "age": {
      "histogram": {
        "field": "age",
        "interval": 2,
        "min_doc_count":1
      }
    }
  }
}

高亮查询 [ highlight ]

# 高亮查询
GET person2/_search
{
  "query": {
    "match": {
      "address": "四川"
    }
  },
  "highlight": {
    "fields": {
      "address": {
         "pre_tags": "<font color='red'>",
          "post_tags": "</font>"
      }
    }
  }
}

批量操作 [ _bulk ]

# 批量操作
# 删除索引为person2，id为5
# 新增索引为person2，id为6
#     内容：名字=徐八，年龄=18，住址=上海市
# 修改索引为person2，id为2
#     内容：名字=刘九
POST _bulk
{"delete":{"_index":"person2","_id":"5"}}
{"create":{"_index":"person2","_id":"6"}}
{"name":"徐八","age":18,"address":"上海市黄浦区"}
{"update":{"_index":"person2","_id":"2"}}
{"doc":{"name":"刘九"}}

查询别名 [ _alias ]

# 查询别名
GET person2/_alias/

起别名

# 起别名
POST person2/_alias/person

复制索引 [ _reindex ]

# 创建索引
PUT person1
{
  "mappings": {
    "properties": {
      "name":{
        "type": "text"
      },
      "age":{
        "type": "integer"
      },
      "address":{
        "type": "keyword"
      }
    }
  }
}
PUT person1/_doc/1
{
  "name":"唐一",
  "age":52,
  "address":"斗罗大陆"
}
GET person1/_search
# person1 拷贝到 person2 中
POST _reindex
{
  "source": {"index": "person1"},
  "dest": {"index": "person2"}
}
# 使用别名查询
GET person/_search

结果

删除别名和索引 [ actions ]

# 删除别名和索引
POST /_aliases
{
  "actions": [
    {
      "remove": {
        "index": "person2",
        "alias": "person"
      }
    }
  ]
}