Elasticsearch spring boot 指定拼音分词器

1下载ik中文/拼音分词器

ik分词器：https://github.com/medcl/elasticsearch-analysis-ik

拼音分词器：https://github.com/medcl/elasticsearch-analysis-pinyin

注意：elasticsearch版本要求严格必须相同

2 安装

1）通过releases找到和es对应版本的zip文件，或者source文件

2）进入elasticsearch安装目录plugins，新建pinyin文件夹

3）将拼音分词器zip文件解压到pinyin目录

4）重启es

3 kibana中配置

1）配置setting

1PUT  my_index
2 {
3        "number_of_shards" : "5",//主分片
4        "number_of_replicas" : "1",//副本
5        "analysis" : {
6          "analyzer" : {
7            "default" : {
8              "tokenizer" : "ik_max_word"//默认多词分词
9            },
10            "pinyin_analyzer" : {
11              "tokenizer" : "my_pinyin"//拼音分词
12            }
13          },
14          "tokenizer" : {
15            //设置拼音分词
16            "my_pinyin" : {
17              "keep_separate_first_letter" : "false",
18              "lowercase" : "true",
19              "type" : "pinyin",
20              "limit_first_letter_length" : "16",
21              "keep_original" : "false",
22              "keep_full_pinyin" : "true"
23            }
24          }
25        }
26      }

2）配置mapping

1PUT my_index/index/_mapping
2{    
3        "properties" : {
4            "name" : {
5                "type" : "keyword",
6                "analyzer" : "ik_max_word",
7                "include_in_all" : true,
8                "fields" : {
9                    "pinyin" : {
10                        "type" : "text",
11                        "analyzer" : "pinyin_analyzer"
12                      }
13                 }
14            }
15      }
16}

4 测试

通过_analyze测试下分词器是否能正常运行：

1GET my_index/_analyze
2{
3"text":"刘德华",
4"analyzer":"pinyin_analyzer"
5}

5 spring boot 中自动创建setting mapping

1）在resources路径下创建usersearch_mapping.json和usersearch_setting.json文件

1usersearch_mapping.json{
2  "index" : {
3    "analysis" : {
4      "analyzer" : {
5        "pinyin_analyzer" : {
6          "tokenizer" : "my_pinyin"
7        }
8      },
9      "tokenizer" : {
10        "my_pinyin" : {
11          "type" : "pinyin",
12          "keep_separate_first_letter" : false,
13          "keep_full_pinyin" : true,
14          "keep_original" : true,
15          "limit_first_letter_length" : 16,
16          "lowercase" : true,
17          "remove_duplicated_term" : true
18        }
19      }
20    }
21  }
22}
23
24usersearch_setting.json
25
26{
27  "user": {
28    "properties": {
29      "title": {
30        "type": "keyword",
31        "fields": {
32          "pinyin": {
33            "type": "text",
34            "store": "no",
35            "term_vector": "with_offsets",
36            "analyzer": "pinyin_analyzer"
37          }
38        }
39      }
40    }
41  }
42}

2）新建测试demo 使用@Mapping和@Setting注解

1@Mapping(mappingPath = "usersearch_setting.json")
2@Setting(settingPath = "usersearch_mapping.json")
3@Document(indexName = "user",type = "user",shards = 5,replicas = 1)
4public class UserIndex  {
5    @Id
6    private  String user;
7//get set省略
8}

3）使用save方法添加数据

使用ElasticsearchTemplate 中的putMapping将setting 和mapping文件执行

1public class UserContrller {
2    @Autowired
3    private UserRepository userRepository;
4    @Autowired
5    private ElasticsearchTemplate elasticsearchTemplate;
6    @RequestMapping("/add")
7    public void add(){
8        //添加配置
9        elasticsearchTemplate.putMapping(User.class);
10        User user =new User();
11        userIndex.setUser("陈奕迅");
12        userRepository.save(user);
13    }
14}