spring-data-elasticsearch suggest实现错词纠正和自动补全

项目中使用了spring-data-elasticsearch注解配置了mapping，实现全文搜索功能。
接到需求需要实现错词纠正功能和自动补全功能：

在网上找的资料，可以通过ES的suggester功能实现，但大部分是直接使用elasticsearch client的方式实现的，于是对使用spring-data-elasticsearch实现的步骤做个总结，希望对有需要的同学有所帮助。
这边直接进行配置说明，不再介绍ES通过注解配置mapping和Suggest的基础知识点。

注意：这里只实现了英文的错词纠正，中文的没有实现*

import io.swagger.annotations.ApiModelProperty;
import lombok.Data;
import lombok.NoArgsConstructor;
import lombok.ToString;
import lombok.experimental.Accessors;
import lombok.experimental.FieldNameConstants;
import org.springframework.data.annotation.Id;
import org.springframework.data.elasticsearch.annotations.*;
import org.springframework.data.elasticsearch.core.completion.Completion;

import java.io.Serializable;
import java.util.List;

@Data
@Accessors(chain = true)
@ToString
@NoArgsConstructor
@FieldNameConstants
@document(indexName = SysProtocol.EsIndexKey.SEARCH_MODEL_EN)
@Setting(settingPath = "/elasticsearch/settings.json")
public class SearchModelEnIndex implements Serializable {
    private static final long serialVersionUID = 1L;

    @Id
    private String id;

    @ApiModelProperty(value = "标题")
    @Field(type = FieldType.Text, analyzer = "my_analyzer")
    private String name;

    @ApiModelProperty(value = "自动补全标题")
    @CompletionField(analyzer = "my_analyzer", searchAnalyzer = "my_analyzer")
    private Completion completionName;
}

我在项目中使用了自定义分词器的配置

{
  "index": {
    "highlight": {
      "max_analyzed_offset": 200000000
    }
  },
  "analysis": {
    "char_filter": {
      "XtoS": {
        "type": "mapping",
        "mappings": ["_=>|",".=>|"]
      }
    },
    "analyzer": {
      "my_analyzer": {
        "type": "custom",
        "char_filter": ["XtoS"],
        "tokenizer": "standard",
        "filter": ["lowercase"]
      }
    }
  }
}

用es suggester的Term实现

TermSuggestionBuilder suggestionBuilder = SuggestBuilders.termSuggestion(SearchModelEnIndex.Fields.name)
			    .suggestMode(TermSuggestionBuilder.SuggestMode.MISSING)
			    .analyzer( "my_analyzer")
			    .size(1); // 设定返回数量
SuggestBuilder suggestBuilder = new SuggestBuilder();
suggestBuilder.addSuggestion("suggest_text", suggestionBuilder).setGlobalText(suggestKeyword);
Suggest suggest = esRestTemplate.suggest(suggestBuilder, SearchModelEnIndex.class).getSuggest();
TermSuggestion suggest_text = suggest.getSuggestion("suggest_text");
for (TermSuggestion.Entry entry : suggest_text.getEntries()) {
    if (entry.getOptions().size() > 0) {
    	System.out.println(entry.getOptions().get(0).getText().toString());
	}
}

用es suggester的Completion实现

		String keyword = "elastic i";
		CompletionSuggestionBuilder completionSuggestionBuilder = SuggestBuilders.completionSuggestion(SearchModelEnIndex.Fields.completionName)
                .size(5)
                .skipDuplicates(true)
                .prefix(keyword);
        SuggestBuilder suggestBuilder = new SuggestBuilder();
        suggestBuilder.addSuggestion("suggest_text", completionSuggestionBuilder);
        Suggest suggest = esRestTemplate.suggest(suggestBuilder, SearchModelEnIndex.class).getSuggest();
        CompletionSuggestion suggest_text = suggest.getSuggestion("suggest_text");
        for (CompletionSuggestion.Entry entry : suggest_text.getEntries()) {
            for (CompletionSuggestion.Entry.Option option : entry.getOptions()) {
    			System.out.println(option.getText().toString());
            }
        }

注意：Completion使用的索引字段类型要用Completion类型：