ELK实现全文模糊搜索高亮

相关项目：后端：gqd000/OurBlog: 多人博客系统 (github.com)

前端：1664635775/BlogDemo (github.com)

后端代码部分：

以搜索文章内容和标题为zhu例，先创建显示的实体类ArticleSearchDTO：

package com.digging.model.dto;

import lombok.AllArgsConstructor;
import lombok.Builder;
import lombok.Data;
import lombok.NoArgsConstructor;
import org.springframework.data.annotation.Id;
import org.springframework.data.elasticsearch.annotations.Document;
import org.springframework.data.elasticsearch.annotations.Field;
import org.springframework.data.elasticsearch.annotations.FieldType;

import java.time.LocalDateTime;

/**
 * 文章搜索结果
 *
 * @author 11921
 */
@Data
@Builder
@AllArgsConstructor
@NoArgsConstructor
@Document(indexName = "article")
public class ArticleSearchDTO {

    /**
     * 文章id
     */
    @Id
    private Long id;

    /**
     * 文章标题
     */
    @Field(type = FieldType.Text, analyzer = "ik_max_word",searchAnalyzer = "ik_smart")
    private String name;

    /**
     * 文章内容
     */
    @Field(type = FieldType.Text, analyzer = "ik_max_word",searchAnalyzer = "ik_smart")
    private String content;

    /**
     * 文章状态
     */
    @Field(type = FieldType.Integer)
    private Integer available;


    //文章作者username
    private String username;
    //文章作者id
    private Long userId;
    //文章创建时间
    private LocalDateTime createTime;
    //文章修改时间
    private LocalDateTime updateTime;

}

ArticleService中加入新方法：

/**
     * 搜索文章
     *
     * @param keywords 条件
     * @return 文章
     */
    List<ArticleSearchDTO> listArticlesBySearch(String keywords);

ArticleServiceImpl加入新方法：

@Override
    public List<ArticleSearchDTO> listArticlesBySearch(String keywords) {
        return searchArticle(buildQuery(keywords));
    }



/**
     * 搜索文章构造
     *
     * @param keywords 条件
     * @return es条件构造器
     */
    private NativeSearchQueryBuilder buildQuery(String keywords) {
        // 条件构造器
        NativeSearchQueryBuilder nativeSearchQueryBuilder = new NativeSearchQueryBuilder();
        BoolQueryBuilder boolQueryBuilder = QueryBuilders.boolQuery();
        // 根据关键词搜索文章标题或内容
        if (Objects.nonNull(keywords)) {
            boolQueryBuilder.must(QueryBuilders.boolQuery().should(QueryBuilders.matchQuery("name", keywords))
                            .should(QueryBuilders.matchQuery("content", keywords)))
                    .must(QueryBuilders.termQuery("available", 0));
        }
        // 查询
        nativeSearchQueryBuilder.withQuery(boolQueryBuilder);
        return nativeSearchQueryBuilder;
    }


    /**
     * 文章搜索结果高亮
     *
     * @param nativeSearchQueryBuilder es条件构造器
     * @return 搜索结果
     */
    private List<ArticleSearchDTO> searchArticle(NativeSearchQueryBuilder nativeSearchQueryBuilder) {
        // 添加文章标题高亮
        HighlightBuilder.Field titleField = new HighlightBuilder.Field("name");
        titleField.preTags("<span style='color:#f47466'>");
        titleField.postTags("</span>");
        // 添加文章内容高亮
        HighlightBuilder.Field contentField = new HighlightBuilder.Field("content");
        contentField.preTags("<span style='color:#f47466'>");
        contentField.postTags("</span>");
        //显示文章全部内容
        contentField.numOfFragments(0);
        //显示除html标签外 200个字符
//        contentField.fragmentSize(200);
        nativeSearchQueryBuilder.withHighlightFields(titleField, contentField);
        // 搜索
        SearchHits<ArticleSearchDTO> search = elasticsearchRestTemplate.search(nativeSearchQueryBuilder.build(), ArticleSearchDTO.class);
        return search.getSearchHits().stream().map(hit -> {
            ArticleSearchDTO article = hit.getContent();
            // 获取文章标题高亮数据
            List<String> titleHighLightList = hit.getHighlightFields().get("name");
            if (CollectionUtils.isNotEmpty(titleHighLightList)) {
                // 替换标题数据
                article.setName(titleHighLightList.get(0));
            }
            // 获取文章内容高亮数据
            List<String> contentHighLightList = hit.getHighlightFields().get("content");
            if (CollectionUtils.isNotEmpty(contentHighLightList)) {
                // 替换内容数据
                article.setContent(contentHighLightList.get(0));
            }
            return article;
        }).collect(Collectors.toList());
    }

ArticleController中编写接口：

//搜索文章
    @GetMapping("/search")
    public Result<List<ArticleSearchDTO>> searchBlog(String keywords)
    {
        List<ArticleSearchDTO> articleSearchDTOS = articleService.listArticlesBySearch(keywords);

        if(articleSearchDTOS.size() == 0)
        {
            return Result.success("未搜索到相关结果！");
        }
        return Result.success(articleSearchDTOS,"搜索成功!");
    }

Logstash配置数据库数据与引擎增量同步

需要对数据库字段更名时，用filter配置：

filter {
    mutate {
    	# 将数据库中“user_id”属性复制到引擎中的“userId字段”
        rename => { "user_id" => "userId" }
        rename => { "create_time" => "createTime" }
        rename => { "update_time" => "updateTime" }  
    }
}

新建article.config文件：

input {
  jdbc {
  	# mysql-connector-java目录，注意版本一致
    jdbc_driver_library => "/home/www/OurBlog/Elastic/Logstash/mysql-connector-java-8.0.24.jar"
    jdbc_driver_class => "com.mysql.cj.jdbc.Driver"
    jdbc_connection_string => "jdbc:mysql://HOST:3306/database"
    jdbc_user => "username"
    jdbc_password => "password"
    # 配置同步时间，各字段含义（由左至右）分、时、天、月、年，全部为*默认含义为每分钟都更新，
    # "*/5 * * * *"代表每5分钟同步一次
    schedule => "* * * * *"
    # 全选，全量更新
    # statement => "SELECT * FROM article"
    # 选择更新时间大于最后一次查询数据库时间条目，增量更新
    statement => "SELECT * FROM article WHERE update_time >= :sql_last_value"
    use_column_value => true
    tracking_column_type => "timestamp"
    tracking_column => "update_time"
    last_run_metadata_path => "syncpoint_table"
  }
}

# 数据库字段重命名再加入到索引
filter {
    mutate {
        rename => { "user_id" => "userId" }
        rename => { "create_time" => "createTime" }
        rename => { "update_time" => "updateTime" }  
    }
}

output {
    elasticsearch {
        # ES的IP地址及端口
        hosts => ["localhost:9200"]
        # 索引名称 可自定义
        index => "article"
        # 需要关联的数据库中有有一个id字段，对应类型中的id
        document_id => "%{id}"
        document_type => "article"
    }
    stdout {
        # JSON格式输出
        codec => json_lines
    }
}

在logstash目录下带着配置文件运行：nohup logstash-7.12.0/bin/logstash -f logstash-7.12.0/config/article.conf &，使用tail -f nohup.out可查看启动日志。