白日梦的ES笔记三：万字长文 Elasticsearch基础概念统一扫盲（二）

PUT /my_index
{
  "settings":{
    "analysis":{
      "analyzer":{
        "es_std":{
          # 指定分词器的类型是：standard
          "type":"standard",
          # 指定分词器的停用词：_english_
          "stopwords":"_english_"
        }
      }
    }
  }
}

PUT my_index
{   
  # 指定index的primary shard数量以及 replicas的数量
  “settings”:{
    "number_of_shards":1,
    "number_of_repicas":0
  },
   # 关键字，我们手动自定my_index中的mapping
  "mappings": {
    "my_index": { # index的名称
      "properties": { # 关键字，mapping的属性，字段
        "my_field1": { # 相当于Java中的   String my_field1
          "type": "text",
           "analyzer":"english"# 指定分词器，说明这个字段需要分词建立倒排索引
        }，
        "my_field2": { # 相当于Golang中的 var my_field2 float
          "type": "float",
          # 指定是否要分词。analyzed表示要，not_analyzed表示不要
          "index":"not_analyzed"
        }，
        "my_field3": {
          "type": "scaled_float"，
          "scaling_factor": 100
        }
      }
    }
  }
}

PUT my_index/_doc/1
{
  "title": "This is a document"
}

# 查看某个index下的某个type的mapping
GET /index/_mapping/type
# 查看某个index的mapping
GET /index/_mapping

PUT my_index/_doc/1
{
  "first_name": "John"
}

null          --> no field add
true flase    --> boolean
123           --> long
123.123       --> double
1999-11-11    --> date
"hello world" --> string
Object        --> object

PUT /my_index/
{
    "mappings":{
        "dynamic":"strict"
    }
}

# 创建mapping并制定：禁用ES的日期探测
PUT my_index
{
  "mappings": {
    "_doc": {
      "date_detection": false
    }
  }
}
# 添加一条doc
PUT my_index/_doc/1 
{
  "create": "1985/12/22"
}
# 查看doc，结果如下
GET my_index/_doc/1 
{
  "_index": "my_index",
  "_type": "_doc",
  "_id": "1",
  "_version": 1,
  "found": true,
  "_source": {
    "create": "1985/12/22"
  }
}
# 查看mapping 
GET my_index/_mapping
# 结果如下：
{
  "my_index": {
    "mappings": {
      "_doc": {
        "date_detection": false,
        "properties": {
          "create": {
            # 被任务是text类型
            "type": "text",
            # ES会自动帮你创建的下面的field部分
            # 即 create是text类型，create.ketword是keyword类型
            # keyword类型不会分词，默认保留前256字符
            "fields": {
              "keyword": {
                "type": "keyword",
                "ignore_above": 256
              }
            }
          }
        }
      }
    }
  }
}

PUT my_index
{
  "mappings": {
    "_doc": {
      "dynamic_date_formats": ["MM/dd/yyyy"]
    }
  }
}
PUT my_index/_doc/1
{
  "create_date": "09/25/2015"
}

PUT my_index
{
  "mappings": {
    "_doc": {
      "numeric_detection": true
    }
  }
}
PUT my_index/_doc/1
{
  "my_float":   "1.0",
  "my_integer": "1" 
}

# 创建index：twitter
PUT twitter
{
  "mappings": {
    # user为type
    "user": {
      "properties": {
        "name": {
         # 会被全部检索
        "type": "text" ,
         # 指定当前field使用 english分词器
        "analyzer":"english" 
        }，
        "user_name": { "type": "keyword" },
        "email": { "type": "keyword" }
      }
    }，
    "tweet": {
      "properties": {
        "content": { "type": "text" },
        "user_name": { "type": "keyword" },
        # "tweeted_at": { "type": "date" },
        "tweeted_at": {
        "type": "date" 
         # 通过index设置为当前field  tweeted_at不能被分词
         "index": "not_analyzeed" 
        }
      }
    }
  }
}

# object类型的json
{
    "address":{
        "province":"shandong",
        "city":"qingdao"
    }，
    "name":"bairimeng",
    "age":"12"
}
# ES会将上面的json转换成如下的格式存储
{
    "name" : [bairimeng],
    "age" : [12],
    "address.province" : [shandong]
    "address.city" : [qingdao]
}

# Object数组类型
{
    "address":[
        {"age":"12"，"name":"张三"},
        {"age":"12"，"name":"张三"},
        {"age":"12"，"name":"张三"}
    ]
}
# ES会将上面的json转换成如下的格式存储
{
    "address.age" : [12，12，12],
    "address.name" : [张三，张三，张三]
}

PUT tweets
{
  "mappings": {
    "_doc": {
      "_source": {
        "enabled": false
      }
    }
  }
}

# Example documents
PUT my_index/_doc/1
{
  "title": "This is a document"
}
PUT my_index/_doc/2?refresh=true
{
  "title": "This is another document",
  "body": "This document has a body"
}

GET my_index/_search
{
  "query": {
    "terms": {
      "_field_names": [ "title" ] 
    }
  }
}

PUT tweets
{
  "mappings": {
    "_doc": {
      "_field_names": {
        "enabled": false
      }
    }
  }
}

# dis_max
GET /your_index/your_type/_search
{   
    # 基于 tie_breaker 优化dis_max
    # tie_breaker可以使dis_max考虑其它field的得分影响
    "query": { 
     # 直接取下面多个query中得分最高的query当成最终得分
     # 这也是best field策略
     "dis_max": { 
        "queries":[
           {"match":{"name":"关注"}},
           {"match":{"content":"白日梦"}}
        ],
        "tie_breaker":0.4
     }
    }
} 
# best_field
# 使用multi_match query简化写法如下：
GET /your_index/your_type/_search
{    
    "query": { 
       "multi_match":{
           "query":"关注 白日梦",
            # 指定检索的策略 best_fields（因为dis_max就是best field策略）
           "type":"best_fields",
            # content^2 表示增加权重，相当于：boost2
           "fields":["name","content^2"],
           "tie_breaker":0.4,
           "minimum_should_match":3
       }
    }
}
# most_field
GET /your_index/your_type/_search
{    
    # most_fields策略、优先返回命中更多关键词的doc
    # 如下从title、name、content中搜索包含“赐我白日梦”的doc
    "query": { 
       "multi_match":{
           "query":"赐我白日梦",
            # 指定检索的策略most_fields
           "type":"most_fields",
           "fields":["title","name","content"]
       }
    }
}

PUT my_index
{
  "mappings": {
    "_doc": {
      "properties": {
        "first_name": {
          "type": "text",
          # 把当前的first_name copy进full_name字段中
          "copy_to": "full_name" 
        }，
        "last_name": {
          "type": "text",
          # 把当前的last_name copy进full_name字段中
          "copy_to": "full_name" 
        }，
        "full_name": {
          "type": "text"
        }
      }
    }
  }
}
PUT my_index/_doc/1
{
  "first_name": "John",
  "last_name": "Smith"
}
GET my_index/_search
{
  "query": {
    "match": {
      "full_name": { 
        "query": "John Smith",
        "operator": "and"
      }
    }
  }
}