elasticsearch分析器三功能自定義

• 字符過濾器(char_filter)

  首先叔收，字符串按順序通過每個(gè)字符過濾器，他們的任務(wù)是在分詞前整理字符串
  一個(gè)字符過濾器可以用來去掉HTML傲隶，或者將&轉(zhuǎn)化成and
  

• 分詞器(tokenizer)

  其次，字符串被分詞器分爲(wèi)單個(gè)的詞條窃页，一個(gè)簡(jiǎn)單的分詞器遇到空格和標(biāo)點(diǎn)的時(shí)候跺株，可能會(huì)將文本拆分成詞條
  Hello how are you?會(huì)被ES預(yù)設(shè)的分詞器standard分成hello、how脖卖、are乒省、you
  

• Token 過濾器 (filter)

  最后，詞條按順序通過每個(gè) token 過濾器畦木，這個(gè)過程可能會(huì)改變?cè)~條(Quick -> quick)袖扛、
  刪除詞條(a、an、and蛆封、the...)唇礁、增加詞條(jump和leap這種同義詞)
  

• 過濾器(filter)解析

  edge_ngram_filter：將每個(gè)詞都進(jìn)行進(jìn)一步的切分，用于即時(shí)搜索(instant search)惨篱。
  `min_gram`表示只要用戶搜索了一個(gè)字符我們就去進(jìn)行匹配盏筐。
  `max_gram`表示匹配的最大長(zhǎng)度，最大長(zhǎng)度越長(zhǎng)越占用空間
  pinyin_simple_filter：拼音首字母的過濾器
  pinyin_full_filter：拼音全拼的過濾器
  

自定義分析器

• 自定義setting格式

  PUT 127.0.0.1:9200/mytest
  {
      "setting": {
          "analysis": {
              "char_filter": { 自定義的字符過濾器 },
              "tokenizer": { 自定義的分詞器 },
              "filter": { 自定義的token過濾器 },
              "analyzer": { 自定義的分析器砸讳，可以將上面的char_filter琢融、tokenizer、filter用不同的組合拼起來簿寂，形成不同的分析器 }
          }
      }
  }
  

實(shí)例

#設(shè)置setting
PUT /enterpriseextend
{
  "settings": {
    "analysis": {
      "filter": {
        "edge_ngram_filter": {
          "type": "edge_ngram",
          "min_gram": 1,
          "max_gram": 50
        },
        "pinyin_simple_filter": {
          "type": "pinyin",
          "keep_first_letter": true,
          "keep_separate_first_letter": false,
          "keep_full_pinyin": false,
          "keep_original": false,
          "limit_first_letter_length": 50,
          "lowercase": true
        },
        "pinyin_full_filter": {
          "type": "pinyin",
          "keep_first_letter": false,
          "keep_separate_first_letter": false,
          "keep_full_pinyin": true,
          "none_chinese_pinyin_tokenize": true,
          "keep_original": false,
          "limit_first_letter_length": 50,
          "lowercase": true
        }
      },
      "tokenizer": {
        "ik_max_word": {
          "type": "ik_max_word",
          "use_smart": true
        }
      },
      "analyzer": {
        "ngramIndexAnalyzer": {
          "type": "custom",
          "tokenizer": "keyword",
          "filter": [
            "edge_ngram_filter",
            "lowercase"
          ]
        },
        "ikIndexAnalyzer": {
          "type": "custom",
          "tokenizer": "ik_max_word"
        },
        "pinyiSimpleIndexAnalyzer": {
          "tokenizer": "keyword",
          "filter": [
            "pinyin_simple_filter",
            "edge_ngram_filter",
            "lowercase"
          ]
        },
        "pinyiFullIndexAnalyzer": {
          "tokenizer": "keyword",
          "filter": [
            "pinyin_full_filter",
            "lowercase"
          ]
        }
      }
    }
  }
}
#設(shè)置mapping
PUT enterpriseextend/_mapping/enterpriseextend
{
  "properties": {
    "id": {
      "type": "long"
    },
    "entName": {
      "type": "text", 
      "analyzer": "ikIndexAnalyzer",
      "fields": {
        "ngram": {
          "type": "text", 
          "analyzer": "ngramIndexAnalyzer"
        },
        "SPY": {
          "type": "text", 
          "analyzer": "pinyiSimpleIndexAnalyzer"
        },
        "FPY": {
          "type": "text", 
          "analyzer": "pinyiFullIndexAnalyzer"
        }
      }
    }
  }
}
#插入語句
PUT enterpriseextend/_doc/1
{
  "entName":"確實(shí)不是啥好東西"
}
#三種查詢
GET enterpriseextend/_search
{
  "query": {
    "match": {
      "entName.ngram": "確實(shí)不是啥好東西"
    }
  }
}
GET enterpriseextend/_search
{
  "query": {
    "match": {
      "entName.SPY": "qsbsshdx"
    }
  }
}
GET enterpriseextend/_search
{
  "query": {
    "match": {
      "entName.FPY": "queshibushishahaodongxi"
    }
  }
}

綜合查詢

GET enterpriseextend/_search
{
  "query": {
    "bool": {
      "should": [
        {
          "match": {
            "entName.ngram": {
              "query": "確實(shí)不是啥好東西",
              "boost": 5 
            }
          }
        },
        {
          "match": { 
            "entName.SPY": {
              "query": "qsbsshdx",
              "boost": 1 
            }
          }
        },
        {
          "match": { 
            "entName.FPY": {
              "query": "queshibushishahaodongxi",
              "boost": 0.8
            }
          }
        }
      ]
    }
  }
}