HTML to DOCX Conversion using Java

html-to-docx-conversion-pandoc.java

import com.github.davidmoten.pandoc.Pandoc;
import java.io.*;

/**
 * Converts HTML business reports to MS Word (DOCX) with corporate styling.
 * Features:
 * - Preserves tables, images, and CSS classes from HTML
 * - Applies custom DOCX template (e.g., company-branded styles)
 * - Handles embedded base64 images
 * - Post-processes with Office-Open-XML (OOXML) hooks
 */
public class HtmlToDocxConverter {

    public static void main(String[] args) {
        // 1. Initialize Pandoc with performance tuning
        Pandoc pandoc = Pandoc.create()
            .memoryLimit(2048); // Increase memory for complex docs

        // 2. Define I/O paths
        String htmlInput = "quarterly_report.html";
        String docxOutput = "Q3_Report.docx";
        String companyTemplate = "templates/corporate_template.docx";

        try (InputStream html = new FileInputStream(htmlInput);
             OutputStream docx = new FileOutputStream(docxOutput)) {

            // 3. Execute conversion with business doc features
            pandoc.from("html")
                  .to("docx")
                  .option("--reference-doc=" + companyTemplate) // Branded template
                  .option("--embed-resources")                  // Inline images/CSS
                  .option("--standalone")                       // Full document
                  .option("--columns=80")                       // Control line wraps
                  .filter("pandoc-secnos")                     // Add section numbers
                  .execute(html, docx);

            System.out.println("DOCX generated: " + docxOutput);

            // 4. Post-processing (e.g., inject live Excel data)
            // Docx4j or Apache POI can modify the DOCX here

        } catch (Exception e) {
            System.err.println("Conversion failed: " + e.getMessage());
            e.printStackTrace();
        }
    }
}