springboot中集成ES怎么实现磁盘文件全文检索功能
本篇内容介绍了“springboot中集成ES怎么实现磁盘文件全文检索功能”的有关知识,在实际案例的操作过程中,不少人都会遇到这样的困境,接下来就让小编带领大家学习一下如何处理这些情况吧!希望大家仔细阅读,能够学有所成!
创新互联建站-专业网站定制、快速模板网站建设、高性价比隆昌网站开发、企业建站全套包干低至880元,成熟完善的模板库,直接使用。一站式隆昌网站制作公司更省心,省钱,快速模板网站建设找我们,业务覆盖隆昌地区。费用合理售后完善,十余年实体公司更值得信赖。
整体架构
考虑到磁盘文件分布到不同的设备上,所以采用磁盘扫瞄代理的模式构建系统,即把扫描服务以代理的方式部署到目标磁盘所在的服务器上,作为定时任务执行,索引统一建立到ES中,当然ES采用分布式高可用部署方法,搜索服务和扫描代理部署到一起来简化架构并实现分布式能力。
磁盘文件快速检索架构
部署ES
ES(elasticsearch)是本项目唯一依赖的第三方软件,ES支持docker方式部署,以下是部署过程
docker pull docker.elastic.co/elasticsearch/elasticsearch:6.3.2 docker run -e ES_JAVA_OPTS="-Xms256m -Xmx256m" -d -p 9200:9200 -p 9300:9300 --name es01 docker.elastic.co/elasticsearch/elasticsearch:6.3.2
部署完成后,通过浏览器打开http://localhost:9200,如果正常打开,出现如下界面,则说明ES部署成功。
ES界面
工程结构
工程结构
依赖包
本项目除了引入springboot的基础starter外,还需要引入ES相关包
org.springframework.boot spring-boot-starter-data-elasticsearch io.searchbox jest 5.3.3 net.sf.jmimemagic jmimemagic 0.1.4
配置文件
需要将ES的访问地址配置到application.yml里边,同时为了简化程序,需要将待扫描磁盘的根目录(index-root)配置进去,后面的扫描任务就会递归遍历该目录下的全部可索引文件。
server: port: @elasticsearch.port@ spring: application: name: @project.artifactId@ profiles: active: dev elasticsearch: jest: uris: http://127.0.0.1:9200 index-root: /Users/crazyicelee/mywokerspace
索引结构数据定义
因为要求文件所在目录、文件名、文件正文都有能够检索,所以要将这些内容都作为索引字段定义,而且添加ES client要求的JestId来注解id。
package com.crazyice.lee.accumulation.search.data; import io.searchbox.annotations.JestId; import lombok.Data; @Data public class Article { @JestId private Integer id; private String author; private String title; private String path; private String content; private String fileFingerprint; }
扫描磁盘并创建索引
因为要扫描指定目录下的全部文件,所以采用递归的方法遍历该目录,并标识已经处理的文件以提升效率,在文件类型识别方面采用两种方式可供选择,一个是文件内容更为精准判断(Magic),一种是以文件扩展名粗略判断。这部分是整个系统的核心组件。
这里有个小技巧
对目标文件内容计算MD5值并作为文件指纹存储到ES的索引字段里边,每次在重建索引的时候判断该MD5是否存在,如果存在就不用重复建立索引了,可以避免文件索引重复,也能避免系统重启后重复遍历文件。
package com.crazyice.lee.accumulation.search.service; import com.alibaba.fastjson.JSONObject; import com.crazyice.lee.accumulation.search.data.Article; import com.crazyice.lee.accumulation.search.utils.Md5CaculateUtil; import io.searchbox.client.JestClient; import io.searchbox.core.Index; import io.searchbox.core.Search; import io.searchbox.core.SearchResult; import lombok.extern.slf4j.Slf4j; import net.sf.jmimemagic.*; import org.apache.poi.hwpf.extractor.WordExtractor; import org.apache.poi.xwpf.extractor.XWPFWordExtractor; import org.apache.poi.xwpf.usermodel.XWPFDocument; import org.elasticsearch.index.query.QueryBuilders; import org.elasticsearch.search.builder.SearchSourceBuilder; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.stereotype.Component; import java.io.File; import java.io.FileInputStream; import java.io.FileNotFoundException; import java.io.IOException; @Component @Slf4j public class DirectoryRecurse { @Autowired private JestClient jestClient; //读取文件内容转换为字符串 private String readToString(File file, String fileType) { StringBuffer result = new StringBuffer(); switch (fileType) { case "text/plain": case "java": case "c": case "cpp": case "txt": try (FileInputStream in = new FileInputStream(file)) { Long filelength = file.length(); byte[] filecontent = new byte[filelength.intValue()]; in.read(filecontent); result.append(new String(filecontent, "utf8")); } catch (FileNotFoundException e) { log.error("{}", e.getLocalizedMessage()); } catch (IOException e) { log.error("{}", e.getLocalizedMessage()); } break; case "doc": //使用HWPF组件中WordExtractor类从Word文档中提取文本或段落 try (FileInputStream in = new FileInputStream(file)) { WordExtractor extractor = new WordExtractor(in); result.append(extractor.getText()); } catch (Exception e) { log.error("{}", e.getLocalizedMessage()); } break; case "docx": try (FileInputStream in = new FileInputStream(file); XWPFDocument doc = new XWPFDocument(in)) { XWPFWordExtractor extractor = new XWPFWordExtractor(doc); result.append(extractor.getText()); } catch (Exception e) { log.error("{}", e.getLocalizedMessage()); } break; } return result.toString(); } //判断是否已经索引 private JSONObject isIndex(File file) { JSONObject result = new JSONObject(); //用MD5生成文件指纹,搜索该指纹是否已经索引 String fileFingerprint = Md5CaculateUtil.getMD5(file); result.put("fileFingerprint", fileFingerprint); SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder(); searchSourceBuilder.query(QueryBuilders.termQuery("fileFingerprint", fileFingerprint)); Search search = new Search.Builder(searchSourceBuilder.toString()).addIndex("diskfile").addType("files").build(); try { //执行 SearchResult searchResult = jestClient.execute(search); if (searchResult.getTotal() > 0) { result.put("isIndex", true); } else { result.put("isIndex", false); } } catch (IOException e) { log.error("{}", e.getLocalizedMessage()); } return result; } //对文件目录及内容创建索引 private void createIndex(File file, String method) { //忽略掉临时文件,以~$起始的文件名 if (file.getName().startsWith("~$")) return; String fileType = null; switch (method) { case "magic": Magic parser = new Magic(); try { MagicMatch match = parser.getMagicMatch(file, false); fileType = match.getMimeType(); } catch (MagicParseException e) { //log.error("{}",e.getLocalizedMessage()); } catch (MagicMatchNotFoundException e) { //log.error("{}",e.getLocalizedMessage()); } catch (MagicException e) { //log.error("{}",e.getLocalizedMessage()); } break; case "ext": String filename = file.getName(); String[] strArray = filename.split("\\."); int suffixIndex = strArray.length - 1; fileType = strArray[suffixIndex]; } switch (fileType) { case "text/plain": case "java": case "c": case "cpp": case "txt": case "doc": case "docx": JSONObject isIndexResult = isIndex(file); log.info("文件名:{},文件类型:{},MD5:{},建立索引:{}", file.getPath(), fileType, isIndexResult.getString("fileFingerprint"), isIndexResult.getBoolean("isIndex")); if (isIndexResult.getBoolean("isIndex")) break; //1. 给ES中索引(保存)一个文档 Article article = new Article(); article.setTitle(file.getName()); article.setAuthor(file.getParent()); article.setPath(file.getPath()); article.setContent(readToString(file, fileType)); article.setFileFingerprint(isIndexResult.getString("fileFingerprint")); //2. 构建一个索引 Index index = new Index.Builder(article).index("diskfile").type("files").build(); try { //3. 执行 if (!jestClient.execute(index).getId().isEmpty()) { log.info("构建索引成功!"); } } catch (IOException e) { log.error("{}", e.getLocalizedMessage()); } break; } } public void find(String pathName) throws IOException { //获取pathName的File对象 File dirFile = new File(pathName); //判断该文件或目录是否存在,不存在时在控制台输出提醒 if (!dirFile.exists()) { log.info("do not exit"); return; } //判断如果不是一个目录,就判断是不是一个文件,时文件则输出文件路径 if (!dirFile.isDirectory()) { if (dirFile.isFile()) { createIndex(dirFile, "ext"); } return; } //获取此目录下的所有文件名与目录名 String[] fileList = dirFile.list(); for (int i = 0; i < fileList.length; i++) { //遍历文件目录 String string = fileList[i]; File file = new File(dirFile.getPath(), string); //如果是一个目录,输出目录名后,进行递归 if (file.isDirectory()) { //递归 find(file.getCanonicalPath()); } else { createIndex(file, "ext"); } } } }
扫描任务
这里采用定时任务的方式来扫描指定目录以实现动态增量创建索引。
package com.crazyice.lee.accumulation.search.service; import lombok.extern.slf4j.Slf4j; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.beans.factory.annotation.Value; import org.springframework.context.annotation.Configuration; import org.springframework.scheduling.annotation.Scheduled; import org.springframework.stereotype.Component; import java.io.IOException; @Configuration @Component @Slf4j public class CreateIndexTask { @Autowired private DirectoryRecurse directoryRecurse; @Value("${index-root}") private String indexRoot; @Scheduled(cron = "* 0/5 * * * ?") private void addIndex(){ try { directoryRecurse.find(indexRoot); directoryRecurse.writeIndexStatus(); } catch (IOException e) { log.error("{}",e.getLocalizedMessage()); } } }
搜索服务
这里以restFul的方式提供搜索服务,将关键字以高亮度模式提供给前端UI,浏览器端可以根据返回的JSON进行展示。
package com.crazyice.lee.accumulation.search.web; import com.alibaba.fastjson.JSONObject; import com.crazyice.lee.accumulation.search.data.Article; import io.searchbox.client.JestClient; import io.searchbox.core.Search; import io.searchbox.core.SearchResult; import io.swagger.annotations.ApiImplicitParam; import io.swagger.annotations.ApiImplicitParams; import io.swagger.annotations.ApiOperation; import lombok.extern.slf4j.Slf4j; import org.elasticsearch.index.query.BoolQueryBuilder; import org.elasticsearch.index.query.QueryBuilders; import org.elasticsearch.search.builder.SearchSourceBuilder; import org.elasticsearch.search.fetch.subphase.highlight.HighlightBuilder; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.lang.NonNull; import org.springframework.web.bind.annotation.PathVariable; import org.springframework.web.bind.annotation.RequestMapping; import org.springframework.web.bind.annotation.RequestMethod; import org.springframework.web.bind.annotation.RestController; import java.io.IOException; import java.util.HashMap; import java.util.List; import java.util.Map; @RestController @Slf4j public class Controller { @Autowired private JestClient jestClient; @RequestMapping(value = "/search/{keyword}",method = RequestMethod.GET) @ApiOperation(value = "全部字段搜索关键字",notes = "es验证") @ApiImplicitParams( @ApiImplicitParam(name = "keyword",value = "全文检索关键字",required = true,paramType = "path",dataType = "String") ) public List search(@PathVariable String keyword){ SearchSourceBuilder searchSourceBuilder = new SearchSourceBuilder(); searchSourceBuilder.query(QueryBuilders.queryStringQuery(keyword)); HighlightBuilder highlightBuilder = new HighlightBuilder(); //path属性高亮度 HighlightBuilder.Field highlightPath = new HighlightBuilder.Field("path"); highlightPath.highlighterType("unified"); highlightBuilder.field(highlightPath); //title字段高亮度 HighlightBuilder.Field highlightTitle = new HighlightBuilder.Field("title"); highlightTitle.highlighterType("unified"); highlightBuilder.field(highlightTitle); //content字段高亮度 HighlightBuilder.Field highlightContent = new HighlightBuilder.Field("content"); highlightContent.highlighterType("unified"); highlightBuilder.field(highlightContent); //高亮度配置生效 searchSourceBuilder.highlighter(highlightBuilder); log.info("搜索条件{}",searchSourceBuilder.toString()); //构建搜索功能 Search search = new Search.Builder(searchSourceBuilder.toString()).addIndex( "gf" ).addType( "news" ).build(); try { //执行 SearchResult result = jestClient.execute( search ); return result.getHits(Article.class); } catch (IOException e) { log.error("{}",e.getLocalizedMessage()); } return null; } }
搜索restFul结果测试
这里以swagger的方式进行API测试。其中keyword是全文检索中要搜索的关键字。
搜索结果
使用thymeleaf生成UI
集成thymeleaf的模板引擎直接将搜索结果以web方式呈现。模板包括主搜索页和搜索结果页,通过@Controller注解及Model对象实现。
“springboot中集成ES怎么实现磁盘文件全文检索功能”的内容就介绍到这里了,感谢大家的阅读。如果想了解更多行业相关的知识可以关注创新互联网站,小编将为大家输出更多高质量的实用文章!
新闻名称:springboot中集成ES怎么实现磁盘文件全文检索功能
标题链接:http://pwwzsj.com/article/phoddi.html