ES拼音插件和es的众多插件一样，简单的说一下操作步骤就是

1，下载(https://github.com/medcl/elasticsearch-analysis-pinyin/releases  选择对应的版本，版本与ES版本一致，建议下载最新版本，然后idea打开适配当前自己使用的es版本号，然后maven打包)。

2，es安装插件，我的es是手动部署到linux服务器上的，es部署下有个目录/plugins/ 新建一个文件夹放进去/plugins/pinyin  然后需要重启es 看日志是否加载成功，当前es更新版本较快，需要适配自己的es版本，否则可能重启不成功。

3，定义新建自己的index索引，很重要，因为我是在已有的项目上去集成的pinyin插件，所以需要支持pinyin插件的同时需要适配之前的业务场景，这里列出es拼音插件的两种配置，希望读者在看到的时候要结合自己的业务逻辑 然后进行调整，拼音插件实现起来简单，但是要做到适合自己还是需要不停的调试

        第一个pinyin插件的参数

keep_first_letter启用此选项时，例如：刘德华> ldh，默认值：true
keep_separate_first_letter启用该选项时，将保留第一个字母分开，例如：刘德华> l，d，h，默认：假的，注意：查询结果也许是太模糊，由于长期过频
limit_first_letter_length 设置first_letter结果的最大长度，默认值：16
keep_full_pinyin当启用该选项，例如：刘德华> [ liu，de，hua]，默认值：true
keep_joined_full_pinyin当启用此选项时，例如：刘德华> [ liudehua]，默认值：false
keep_none_chinese 在结果中保留非中文字母或数字，默认值：true
keep_none_chinese_together保持非中国信一起，默认值：true，如：DJ音乐家- > DJ，yin，yue，jia，当设置为false，例如：DJ音乐家- > D，J，yin，yue，jia，注意：keep_none_chinese必须先启动
keep_none_chinese_in_first_letter第一个字母保持非中文字母，例如：刘德华AT2016- > ldhat2016，默认值：true
keep_none_chinese_in_joined_full_pinyin保留非中文字母加入完整拼音，例如：刘德华2016- > liudehua2016，默认：false
none_chinese_pinyin_tokenize打破非中国信成单独的拼音项，如果他们拼音，默认值：true，如：liudehuaalibaba13zhuanghan- > liu，de，hua，a，li，ba，ba，13，zhuang，han，注意：keep_none_chinese和keep_none_chinese_together应首先启用
keep_original 当启用此选项时，也会保留原始输入，默认值：false
lowercase 小写非中文字母，默认值：true
trim_whitespace 默认值：true
remove_duplicated_term当启用此选项时，将删除重复项以保存索引，例如：de的> de，默认值：false，注意：位置相关查询可能受影响

这个需要调试调到自己合适的参数值：业务场景是结合IK分词器分出刘德华 -> liudehua ,刘德华（保留用户输入）

"full_pinyin" : {
              "keep_joined_full_pinyin" : "true",
              "keep_none_chinese_in_first_letter" : "false",
              "lowercase" : "true",
              "none_chinese_pinyin_tokenize" : "false",
              "keep_none_chinese_in_joined_full_pinyin" : "true",
              "keep_original" : "true",
              "keep_first_letter" : "false",
              "keep_separate_first_letter" : "false",
              "type" : "pinyin",
              "keep_none_chinese" : "true",
              "limit_first_letter_length" : "16",
              "keep_full_pinyin" : "false"
            }

这个需要调试调到自己合适的参数值：业务场景是结合IK分词器分出刘德华 -> liudehua 

"full_pinyin_letter" : {
              "keep_joined_full_pinyin" : "true",
              "keep_none_chinese_in_first_letter" : "false",
              "lowercase" : "true",
              "none_chinese_pinyin_tokenize" : "false",
              "keep_none_chinese_in_joined_full_pinyin" : "true",
              "keep_original" : "false",
              "keep_first_letter" : "false",
              "keep_separate_first_letter" : "false",
              "type" : "pinyin",
              "keep_none_chinese" : "true",
              "limit_first_letter_length" : "16",
              "keep_full_pinyin" : "false"
            }

这两种能适配我现在的业务场景

pinyin的实现方式有很多种，这里介绍我自己的实现

        pinyin结合自动补全completion suggester 自动补全

这里用的分析器的实现 还有一种用 filter的实现

  "suggest_v1" : {
    "settings" : {
      "index" : {
        "number_of_shards" : "1",
        "blocks" : {
          "read_only_allow_delete" : "false"
        },
        "provided_name" : "suggest_v1",
        "max_result_window" : "1000000",
        "creation_date" : "164444429959",
        "analysis" : {
          "analyzer" : {
            "test_ik" : {
              "filter" : [
                "lowercase"
              ],
              "char_filter" : [
                "html_strip"
              ],
              "tokenizer" : "ik_max_word"
            },
            "test_ik_search" : {
              "filter" : [
                "lowercase"
              ],
              "char_filter" : [
                "html_strip"
              ],
              "tokenizer" : "ik_smart"
            },
            "pinyin_analyzer" : {
              "tokenizer" : "pinyin_letter"
            }
          },
          "tokenizer" : {
            "pinyin_letter" : {
              "keep_joined_full_pinyin" : "true",
              "keep_none_chinese_in_first_letter" : "false",
              "lowercase" : "true",
              "none_chinese_pinyin_tokenize" : "false",
              "keep_none_chinese_in_joined_full_pinyin" : "true",
              "keep_original" : "false",
              "keep_first_letter" : "false",
              "keep_separate_first_letter" : "false",
              "type" : "pinyin",
              "keep_none_chinese" : "true",
              "limit_first_letter_length" : "16",
              "keep_full_pinyin" : "false"
            }
          }
        }
      
      }
    }
  }

"mappings" : {
      "dynamic" : "false",
      "properties" : {
        "qqqq" : {
          "type" : "text",
          "fields" : {
            "keyword" : {
              "type" : "keyword"
            },
            "pinyin_suggest" : {
              "type" : "completion",
              "analyzer" : "pinyin_analyzer",
              "preserve_separators" : true,
              "preserve_position_increments" : true,
              "max_input_length" : 50
            }
          },
          "analyzer" : "test_ik",
          "search_analyzer" : "test_ik_search"
        },
        "aaaasearch" : {
          "type" : "text",
          "fields" : {
            "keyword" : {
              "type" : "keyword"
            },
            "pinyin_suggest" : {
              "type" : "completion",
              "analyzer" : "pinyin_analyzer",
              "preserve_separators" : true,
              "preserve_position_increments" : true,
              "max_input_length" : 50
            }
          },
          "analyzer" : "test_ik",
          "search_analyzer" : "test_ik_search"
        },
        "title" : {
          "type" : "text",
          "fields" : {
            "keyword" : {
              "type" : "keyword"
            },
            "pinyin_suggest" : {
              "type" : "completion",
              "analyzer" : "pinyin_analyzer",
              "preserve_separators" : true,
              "preserve_position_increments" : true,
              "max_input_length" : 50
            }
          },
          "analyzer" : "test_ik",
          "search_analyzer" : "test_ik_search"
        }
      }
    }

然后使用completion suggester 前缀匹配即可实现 搜索建议

第二种使用filter

analysis" : {
          "filter" : {
            "full_pinyin" : {
              "keep_joined_full_pinyin" : "true",
              "keep_none_chinese_in_first_letter" : "false",
              "lowercase" : "true",
              "none_chinese_pinyin_tokenize" : "false",
              "keep_none_chinese_in_joined_full_pinyin" : "true",
              "keep_original" : "true",
              "keep_first_letter" : "false",
              "keep_separate_first_letter" : "false",
              "type" : "pinyin",
              "keep_none_chinese" : "true",
              "limit_first_letter_length" : "16",
              "keep_full_pinyin" : "false"
            }
          },
          "analyzer" : {
            "test_ik" : {
              "filter" : [
                "lowercase",
               
                "full_pinyin"
              ],
              "char_filter" : [
                "html_strip"
              ],
              "tokenizer" : "ik_max_word"
            },
            "test_ik_search" : {
              "filter" : [
                "lowercase",
                
                "full_pinyin"
              ],
              "char_filter" : [
                "html_strip"
              ],
              "tokenizer" : "ik_smart"
            }
          }
        }

都是结合ik实现