Skip to content

Latest commit

 

History

History
234 lines (178 loc) · 4.34 KB

indexing_data.md

File metadata and controls

234 lines (178 loc) · 4.34 KB

Creating indices

Elasticsearch会自动创建index，譬如

http POST :9200/blog/article/1 title="hello world"

如果blog这个index刚开始不存在，Elasticsearch会自动创建。但我们也能够通过如下方式手动创建:

http PUT :9200/blog

HTTP/1.1 200 OK
Content-Length: 21
Content-Type: application/json; charset=UTF-8

{
    "acknowledged": true
}

创建的时候指定一些配置

http PUT :9200/blog settings:='{"number_of_shards":1, "number_of_replicas": 2}'
HTTP/1.1 200 OK
Content-Length: 21
Content-Type: application/json; charset=UTF-8

{
    "acknowledged": true
}

上面我们创建了一个blog索引，使用一个shard，两个replicas。

http GET :9200/blog

HTTP/1.1 200 OK
Content-Length: 191
Content-Type: application/json; charset=UTF-8

{
    "blog": {
        "mappings": {}, 
        "settings": {
            "index": {
                "creation_date": "1421133862199", 
                "number_of_replicas": "2", 
                "number_of_shards": "1", 
                "uuid": "Vf0y-AZ-Q8uU26u5ptchkQ", 
                "version": {
                    "created": "1040299"
                }
            }
        }
    }
}

我们也能很方便的删除index

http DELETE :9200/blog

HTTP/1.1 200 OK
Content-Length: 21
Content-Type: application/json; charset=UTF-8

{
    "acknowledged": true
}

Mappings

Schema Mappings可以认为是关系型数据库中的schema，它会定义index的相关结构。

下面以一个posts index为例，一个blog post，包含如下结构:

  • Unique identifier
  • Name
  • Publication date
  • Contents

对应的mapping文件, posts.json:

{
    "mappings": {
        "post": {
            "properties": {
                "id" : {"type":"long", "store":"yes"},
                "name" : {"type":"string", "store":"yes", "index":"analyzed"},
                "published" : {"type":"date", "store":"yes"},
                "contents" : {"type":"long", "store":"no", "index":"analyzed"}
            }
        }
    }
}

http POST :9200/posts < posts.json 

HTTP/1.1 200 OK
Content-Length: 21
Content-Type: application/json; charset=UTF-8

{
    "acknowledged": true
}


http GET :9200/posts 

HTTP/1.1 200 OK
Content-Length: 402
Content-Type: application/json; charset=UTF-8

{
    "posts": {
        "mappings": {
            "post": {
                "properties": {
                    "contents": {
                        "index": "analyzed", 
                        "type": "long"
                    }, 
                    "id": {
                        "store": true, 
                        "type": "long"
                    }, 
                    "name": {
                        "store": true, 
                        "type": "string"
                    }, 
                    "published": {
                        "format": "dateOptionalTime", 
                        "store": true, 
                        "type": "date"
                    }
                }
            }
        },
    ...
}

Core types

  • String
  • Number
  • Date
  • Boolean
  • Binary

Common attributes

  • index_name
  • index：analyzed，这个field就会被索引并且能够搜索，no，不能被索引，如果是string，则还有not_analyzed，表明能被索引，不能被analyzed，只能完全匹配
  • store：yes，原始值被写入index，no，不写入，默认为no
  • boost：默认为1，值越高，表明这个field越重要
  • null_value
  • copy_to
  • include_in_all

String

  • term_vector
  • omit_norms
  • analyzer：定义indexing和searching使用的analyzer
  • index_analyzer
  • serach_analyzer
  • norms.enabled
  • norms.loading
  • position_offset_gap
  • index_options
  • ignore_above

Number

包括byte, short, integer, long, float, double

  • precision_step
  • ignore_malformed

Boolean

true or false

Binary

在index使用base64编码存储，只有index_name属性

Date

  • format：日期格式，如"YYYY-mm-dd"
  • precision_step
  • ignore_malformed

Multi fields

"name" : {
    "type" : "string",
    "fields" : {
        "facet" : {"type" : "string", "index":"not_analyzed"}
    }
}

IP Address

"address" : {"type" : "ip", "store" : "yes"}

token_count

"address_count" : {"type" : "token_count", "store" : "yes"}

Analyzers

使用ik进行中文分词处理