SpringBoot + Apache tika 轻松实现各种文档内容解析

戳上方蓝字“Java笔记虾”关注！

Apache tika是Apache开源的一个文档解析工具。Apache Tika可以解析和提取一千多种不同的文件类型(如PPT、XLS和PDF)的内容和格式，并且Apache Tika提供了多种使用方式，既可以使用图形化操作页面（tika-app），又可以独立部署（tika-server）通过接口调用，还可以引入到项目中使用。

本文演示在Spring boot 中引入tika的方式解析文档。如下：

引入依赖

在spring boot 项目中引入如下依赖:

  <dependencyManagement>
    <dependencies>
      <dependency>
        <groupId>org.apache.tika</groupId>
        <artifactId>tika-bom</artifactId>
        <version>2.8.0</version>
        <type>pom</type>
        <scope>import</scope>
      </dependency>
    </dependencies>
  </dependencyManagement>

    <dependency>
      <groupId>org.apache.tika</groupId>
      <artifactId>tika-core</artifactId>
    </dependency>
    <dependency>
      <groupId>org.apache.tika</groupId>
      <artifactId>tika-parsers-standard-package</artifactId>
    </dependency>

创建配置

将tika-config.xml文件放在resources目录下。tika-config.xml文件的内容如下：

<?xml version="1.0" encoding="UTF-8"?>
<properties>
    <encodingDetectors>
        <encodingDetector class="org.apache.tika.parser.html.HtmlEncodingDetector">
            <params>
                <param name="markLimit" type="int">64000</param>
            </params>
        </encodingDetector>
        <encodingDetector class="org.apache.tika.parser.txt.UniversalEncodingDetector">
            <params>
                <param name="markLimit" type="int">64001</param>
            </params>
        </encodingDetector>
        <encodingDetector class="org.apache.tika.parser.txt.Icu4jEncodingDetector">
            <params>
                <param name="markLimit" type="int">64002</param>
            </params>
        </encodingDetector>
    </encodingDetectors>
</properties>

创建配置类MyTikaConfig

import java.io.IOException;
import java.io.InputStream;
import org.apache.tika.Tika;
import org.apache.tika.config.TikaConfig;
import org.apache.tika.detect.Detector;
import org.apache.tika.exception.TikaException;
import org.apache.tika.parser.AutoDetectParser;
import org.apache.tika.parser.Parser;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;
import org.springframework.core.io.Resource;
import org.springframework.core.io.ResourceLoader;
import org.xml.sax.SAXException;

/**
 * tika配置类
 */
@Configuration
public class MyTikaConfig {

    @Autowired
    private ResourceLoader resourceLoader;

    @Bean
    public Tika tika() throws TikaException, IOException, SAXException {

        Resource resource = resourceLoader.getResource("classpath:tika-config.xml");
        InputStream inputStream = resource.getInputStream();

        TikaConfig config = new TikaConfig(inputStream);
        Detector detector = config.getDetector();
        Parser autoDetectParser = new AutoDetectParser(config);

        return new Tika(detector, autoDetectParser);
    }
}