基于Java实现Word转EPUB的实现方案详解

<repositories>
    <repository>
        <id>com.e-iceblue</id>
        <name>e-iceblue</name>
        <url>https://repo.e-iceblue.cn/repository/maven-public/</url>
    </repository>
</repositories>
<dependencies>
    <dependency>
        <groupId>e-iceblue</groupId>
        <artifactId>spire.doc</artifactId>
        <version>14.4.0</version>
    </dependency>
</dependencies>

import com.spire.doc.Document;
import com.spire.doc.FileFormat;

public class ConvertWordToEpub {
    public static void main(String[] args) {
        // 创建 Document 实例
        Document doc = new Document();
        
        // 加载 Word 文档
        doc.loadFromFile("Sample.docx");
        
        // 保存为 EPUB 格式
        doc.saveToFile("ToEpub.epub", FileFormat.E_Pub);
    }
}

import com.spire.doc.Document;
import com.spire.doc.fields.DocPicture;

public class ConvertWordToEpubWithCover {
    public static void main(String[] args) {
        // 加载 Word 文档
        Document doc = new Document();
        doc.loadFromFile("Sample.docx");
        
        // 创建图片对象并加载封面图
        DocPicture coverPicture = new DocPicture(doc);
        coverPicture.loadImage("cover.png");
        
        // 保存时附带封面图片
        doc.saveToEpub("WithCover.epub", coverPicture);
    }
}

<repositories>
    <repository>
        <id>aspose-repo</id>
        <url>https://releases.aspose.com/java/repo/</url>
    </repository>
</repositories>
<dependency>
    <groupId>com.aspose</groupId>
    <artifactId>aspose-words</artifactId>
    <version>24.4</version>
    <classifier>jdk17</classifier>
</dependency>

import com.aspose.words.*;
public class WordToEpub {
    public static void main(String[] args) throws Exception {
        Document doc = new Document("input.docx");
        doc.save("output.epub");
    }
}

import com.aspose.words.*;
public class WordToEpubAdvanced {
    public static void main(String[] args) throws Exception {
        Document doc = new Document("input.docx");
        // 配置 EPUB 保存选项
        HtmlSaveOptions options = new HtmlSaveOptions();
        options.setSaveFormat(SaveFormat.EPUB);
        options.setEncoding(StandardCharsets.UTF_8);
        // 按标题段落拆分（大型文档拆分为多个章节）
        options.setDocumentSplitCriteria(DocumentSplitCriteria.HEADING_PARAGRAPH);
        // 导出文档属性（作者、创建日期等元数据）
        options.setExportDocumentProperties(true);
        doc.save("output.epub", options);
    }
}

<dependency>
    <groupId>org.docx4j</groupId>
    <artifactId>docx4j-JAXB-Internal</artifactId>
    <version>8.0.0</version>
</dependency>

<repositories>
    <repository>
        <id>psiegman-repo</id>
        <url>https://github.com/psiegman/mvn-repo/raw/master/releases</url>
    </repository>
</repositories>
<dependency>
    <groupId>nl.siegmann.epublib</groupId>
    <artifactId>epublib-core</artifactId>
    <version>3.1</version>
</dependency>

import org.docx4j.convert.out.HTMLSettings;
import org.docx4j.openpackaging.packages.WordprocessingMLPackage;
import nl.siegmann.epublib.domain.Book;
import nl.siegmann.epublib.domain.Author;
import nl.siegmann.epublib.domain.Resource;
import nl.siegmann.epublib.epub.EpubWriter;
public class Docx4jToEpub {
    public static void main(String[] args) throws Exception {
        // 1. docx4j 加载 Word 并转换为 HTML 字符串
        WordprocessingMLPackage wordMLPackage = 
            WordprocessingMLPackage.load(new File("input.docx"));
        HTMLSettings settings = new HTMLSettings();
        String htmlContent = wordMLPackage.getHTMLPage(settings);
        // 2. Epublib 构建 EPUB
        Book book = new Book();
        book.getMetadata().addTitle("Converted Document");
        book.getMetadata().addAuthor(new Author("Author", "Name"));
        book.getMetadata().setCoverImage(new Resource(...));
        // 3. 添加 HTML 内容作为章节
        book.addSection("Main Content", new Resource(htmlContent.getBytes(), "content.html"));
        // 4. 保存 EPUB
        EpubWriter writer = new EpubWriter();
        writer.write(book, new FileOutputStream("output.epub"));
    }
}