ElasticSearch学习（四）—— 中文按拼音排序&拼音检索

你的名字 2023-02-27 05:43 27阅读 0赞

*  使用pinyin分词器

1.  下载拼音分词器https://github.com/medcl/elasticsearch-analysis-pinyin
2.  解压进入目录elasticsearch-analysis-pinyin
3.  修改pom.xml中es版本为自己使用的版本
4.  命令行mvn package 打包
5.  进入elasticsearch-analysis-pinyin-master\\target\\releases解压elasticsearch-analysis-pinyin-7.7.0.zip文件
6.  将解压后的文件拷贝到es安装目录下的 plugins/pinyin 中
7.  重启es
8.  修改报错的地方，重复步骤4567

*  索引设置

PUT /book
    {
        "settings": {
            "index": {
                "analysis": {
                    "analyzer": {
                        "pinyin_analyzer": {
                            "tokenizer": "my_pinyin"
                        }
                    }, 
                    "tokenizer": {
                        "my_pinyin": {
                            "type": "pinyin", 
                            "keep_none_chinese": false,
                            "keep_full_pinyin": false,
                            "keep_joined_full_pinyin": true,
                            "keep_none_chinese_in_joined_full_pinyin": true,
                            "keep_first_letter": false,
                            "keep_none_chinese_in_first_letter": false,
                            "none_chinese_pinyin_tokenize": false
                        }
                    }
                }
            }
        }
    }

'keep\_none\_chinese' => false, // 对非中文不拆分词

'keep\_full\_pinyin' => false, // 关闭: 刘德华 -> liu, de, hua

'keep\_joined\_full\_pinyin' => true, // 刘德华 -> liudehua

'keep\_none\_chinese\_in\_joined\_full\_pinyin' => true, // 刘德华2016 -> liudehua2016

'keep\_first\_letter' => true, // 刘德华 -> ldh

'keep\_none\_chinese\_in\_first\_letter' => true, // 刘德华2016 -> ldh2016

'none\_chinese\_pinyin\_tokenize' => false, // 没有卵用

keep\_separate\_first\_letter :将字母分割，例如：刘德华> l，d，h，default：false。

keep\_full\_pinyin :包含全拼拼音，例如：刘德华> \[ liu，de，hua\]，default：true。

limit\_first\_letter\_length :设置first\_letter结果的最大长度，default：16。

lowercase :小写非中文字母，default：true。

keep\_none\_chinese : 不在结果中保留非中文字母或数字，default：true。

*   设置字段

POST /book/_mapping
    {
        "properties": {
            "title": {
              "type": "text",                              
                     "fields": {                                 
                        "keyword": {                            
                          "type": "keyword",
                          "ignore_above": 256
                        },
                        "sort": {
                          "type": "text",
                          "analyzer": "pinyin_analyzer"
                        }
                    }
            },
            "author": {                                        
                "type": "text",                              
                "fields": {                                 
                     "keyword": {                            
                        "type": "keyword",
                        "ignore_above": 256
                    },
                  "sort": {
                    "type": "text",
                    "analyzer": "pinyin_analyzer"
                  }
                }
            }
        }
    }

注意：Only [text][] fields support the analyzer mapping parameter.只有text可以设置分词器

*   检索

GET /book/_search
    {
      "query": {
        "match": {
          "title": "测试"
        }
      },
      "from": 0, 
      "size": 20,
      "sort": {
          "title.sort" : "asc"
        }
    }

*  拼音检索
    
     *  安装ik分词器步骤[https://github.com/medcl/elasticsearch-analysis-ik][https_github.com_medcl_elasticsearch-analysis-ik]同拼音分词器
     *  安装完成需要重启es
     *  索引设置

PUT /book
    {
        "settings": {
            "index": {
                "analysis": {
                    "analyzer": {
                        "ik_smart_pinyin": {
                            "type": "custom",
                            "tokenizer": "ik_smart",
                            "filter": "my_pinyin_filter"
                        },
                        "ik_max_word_pinyin": {
                            "type": "custom",
                            "tokenizer": "ik_max_word",
                            "filter": "my_pinyin_filter"
                        },
                        "pinyin_analyzer": {
                            "tokenizer": "my_pinyin_tokenizer"
                        }
                    }, 
                    "tokenizer": {
                        "my_pinyin_tokenizer": {
                            "type": "pinyin", 
                            "keep_first_letter": false, 
                            "keep_full_pinyin": false, 
                            "keep_joined_full_pinyin": true, 
                            "keep_none_chinese_in_first_letter": true, 
                            "none_chinese_pinyin_tokenize": false, 
                            "lowercase": true, 
                            "with_tone_number": true
                        }
                    },
                    "filter": {
                      "my_pinyin_filter": {
                            "type": "pinyin", 
                            "keep_first_letter": false, 
                            "keep_full_pinyin": false, 
                            "keep_joined_full_pinyin": true, 
                            "keep_none_chinese_in_first_letter": true, 
                            "none_chinese_pinyin_tokenize": false, 
                            "lowercase": true, 
                            "with_tone_number": true
                        }
                    }
                }
            }
        }
    }

*  字段设置

POST /ancientbook/_mapping
    {
        "properties": {
            "title": {                                        
                    "type": "text",                     
                    "analyzer": "ik_max_word_pinyin",          
                     "fields": {                        
                         "keyword": {                 
                            "type": "keyword",
                            "ignore_above": 256
                        }
                    }
                },
                "author": {                                       
                    "type": "text",                     
                    "analyzer": "ik_max_word_pinyin",          
                     "fields": {                       
                         "keyword": {                   
                            "type": "keyword",
                            "ignore_above": 256
                        }
                    }
                }
        }
    }

该配置可以实现中文/拼音检索，但无法按拼音排序。

*  使用icu分词器
    
     *  安装插件
        
         *  es安装目录下查看插件./bin/elasticsearch-plugin list
         *  es安装目录下./bin/elasticsearch-plugin install analysis-icu
         *  重启es
     *  索引设置

PUT /book
    
    POST /book/_mapping
    {
        "properties": {
            "title": {                                        
                    "type": "text",                     
                    "analyzer": "icu_analyzer",  
                     "fields": {                        
                         "keyword": {                 
                            "type": "keyword",
                            "ignore_above": 256
                        },
                      "sort": {  
                        "type": "icu_collation_keyword",
                        "index": false,
                        "language": "zh",
                        "country": "CN"
                      }
                    }
                },
                "author": {                                       
                    "type": "text",                     
                    "analyzer": "icu_analyzer",          
                     "fields": {                       
                         "keyword": {                   
                            "type": "keyword",
                            "ignore_above": 256
                        },
                      "sort": {  
                        "type": "icu_collation_keyword",
                        "index": false,
                        "language": "zh",
                        "country": "CN"
                      }
                    }
                }
        }
    }

*  检索

GET /book/_search
    {
        "query": {
            "bool": {
                "must": [
                    {
                        "match_all": {
                          
                        }
                    }
                ]
            }
        }, 
        "from": 0, 
        "size": 10, 
        "sort": [
            {
                "title.sort": "asc"
            }
        ]
    }

*  同时需要分词检索和按拼音排序使用的是icu分词器

1.es安装目录下查看插件./bin/elasticsearch-plugin list
    2.es安装目录下./bin/elasticsearch-plugin install analysis-icu
    3.重启es

[text]: https://www.elastic.co/guide/en/elasticsearch/reference/current/text.html
[https_github.com_medcl_elasticsearch-analysis-ik]: https://github.com/medcl/elasticsearch-analysis-ik