索引映射

习题1

创建一个索引，符合如下的要求：

• 索引名称为movies
• 同时为索引创建一个相应的alias，通过此别名查询的数据，name字段都是以movie_开头的。
• 索引的主分片数量为3，为每个主分片分配一个副本。
• 使用bulk API，对索引movies写入多条电影数据。

答案：

# 新增索引直接指定别名
PUT movies
{
  "mappings": {
    "properties": {
      "name":{
        "type": "text",
        "fields": {
          "keyword": {
            "type": "keyword",
            "ignore_above": 256
          }
        }
      }
    }
  }, 
  "settings": {
    "number_of_shards": 3,
    "number_of_replicas": 1
    
  }, 
  "aliases": {
    "movies_a": {
      "filter": {
        "prefix": {
          "name": "movie_"
        }
      }
    }
  }
}

POST movies_a/_bulk
{"index": {"_id": 1}}
{"name": "123123"}
{"index": {"_id": 2}}
{"name": "movie_1"}
{"index": {"_id": 3}}
{"name": "movie_2"}
{"index": {"_id": 4}}
{"name": "xxxxx"}

GET movies/_search

GET movies_a/_search

习题2

查询满足条件的省级行政单位信息，并且满足以下条件之一：

• 城市信息中包含北京市
• 行政区信息中包含淇滨区

以下为习题所需数据：

PUT /area/_doc/1
{
  "province": {
    "name": "北京",
    "cities": [
      {
        "name": "北京市",
        "district": [
          {"name":"丰台区"},
          {"name":"海淀区"},
          {"name":"朝阳区"},
          {"name":"东城区"},
          {"name":"西城区"},
          {"name":"昌平区"}
          ]
      }
    ]
  }
}
PUT /area/_doc/2
{
  "province": {
    "name": "河南省",
    "cities": [
      {
        "name": "郑州市",
        "district": [
          {
            "name": "金水区"
          },
          {
            "name": "高新区"
          },
          {
            "name": "郑东新区"
          },
          {
            "name": "二七区"
          },
          {
            "name": "中原区"
          },
          {
            "name": "惠济区"
          }
        ]
      },
      {
        "name": "鹤壁市",
        "district": [
          {
            "name": "山城区"
          },
          {
            "name": "淇滨区"
          },
          {
            "name": "鹤山区"
          },
          {
            "name": "朝歌"
          },
          {
            "name": "浚县"
          }
        ]
      }
    ]
  }
}
PUT /area/_doc/3
{
  "province": {
    "name": "台湾省",
    "cities": [
      {
        "name": "台北市",
        "district": [
          {
            "name": "中正区"
          },
          {
            "name": "大同区"
          },
          {
            "name": "中山区"
          },
          {
            "name": "万华区"
          },
          {
            "name": "信义区"
          },
          {
            "name": "松山区"
          }
        ]
      },
      {
        "name": "高雄",
        "district": [
          {
            "name": "小港区"
          },
          {
            "name": "鼓山区"
          },
          {
            "name": "三民区"
          }
        ]
      }
    ]
  }
}


答案：
1. 先指定mapping nested对象类型，针对查询的name字段将text类型换成keyword不分词
PUT area
{
  "mappings": {
    "properties": {
      "province": {
        "type": "nested",
        "properties": {
          "name": {
            "type": "keyword"
          },
          "cities": {
            "type": "nested",
            "properties": {
              "name": {
                "type": "keyword"
              },
              "district": {
                "type": "nested",
                "properties": {
                  "name": {
                    "type": "keyword"
                  }
                }
              }
            }
          }
        }
      }
    }
  }
}


2. 执行查询语句  这里按照两个对象nested类型 进行查询，采用的是 bool should，只要一下两个条件满足任意就直接查询出来
GET area/_search
{
  "query": {
    "bool": {
      "should": [
        {
          "nested": {
            "path": "province.cities",
            "query": {
              "term": {
                "province.cities.name": "北京市"
              }
            }
          }
        },
        {
          "nested": {
            "path": "province.cities.district",
            "query": {
              "term": {
                "province.cities.district.name": "淇滨区"
              }
            }
          }
        }
      ]
    }
  }
}

习题3

为索引task3，设置dynamic mapping，具体的要求如下：

一切text类型的字段，类型全部映射为keyword
一切以double_开头命名的字段，类型都设置成double
bulk导入如下的数据进行验证。
导入数据

POST task3/_bulk
{"index":{"_id":1}}
{"cont":"你好马士兵教育", "int_value":35}
{"index":{"_id":2}}
{"cont":"马士兵教育", "int_value":35}
{"index":{"_id":3}}
{"cont":"马士兵", "int_value":35}

答案：

PUT task3
{
  "mappings": {
    "dynamic_templates": [
      {
        "text_2_keyword": {
          "match_mapping_type": "string",
          "mapping": {
            "type": "keyword"
          }
        }
      },
      {
        "int_prefix_2_integer": {
          "match": "double_*",
          "mapping": {
            "type": "double"
          }
        }
      }
    ]
  }
}

#PUT _template/day3_task3_dynamic_template
#{
#  "index_patterns": ["task3"],
#  "mappings": {
#    "dynamic_templates": [
#      {
#        "text_2_keyword": {
#          "match_mapping_type": "string",
#          "mapping": {
#            "type": "keyword"
#          }
#        }
#      },
#      {
#        "int_prefix_2_integer": {
#          "match": "double_*",
#          "mapping": {
#            "type": "double"
#          }
#        }
#      }
#    ]
#  }
#}

验证：
GET task3/_mapping
{
  "task3" : {
    "mappings" : {
      "dynamic_templates" : [
        {
          "strings_as_keyword" : {
            "match_mapping_type" : "string",
            "runtime" : {
              "type" : "keyword"
            }
          }
        },
        {
          "strings_as_int" : {
            "match" : "double_*",
            "runtime" : {
              "type" : "double"
            }
          }
        }
      ],
      "runtime" : {
        "cont" : {
          "type" : "keyword"
        },
        "int_value" : {
          "type" : "double"
        }
      }
    }
  }
}

习题4

设置一个index template，符合如下的要求：

为msb和msb-开头的索引，创建3个主分片，1个副本分片。
对索引msb-tech写入测试数据


答案：

PUT _index_template/day3_task4_index_template
{
  "index_patterns": ["msb", "msb-*"],
  "template": {
    "settings": {
      "number_of_shards": 3,
      "number_of_replicas": 1
    }
  }
}

PUT msb-tech/_doc/1
{
  "title": "测试数据1"
}

GET _cat/indices?v

结果 pri=3 rep=1