Microsoft Office文書を画像に変換するベストオープンソースAPI

sudo apt-get update
sudo apt-get install -y libreoffice-headless poppler-utils

const { exec } = require('child_process');
const path = require('path');
const fs = require('fs');

function convertDocToImage(inputDocPath, outputDir) {
    return new Promise((resolve, reject) => {
        // Step 1: Headless LibreOffice conversion to PDF
        const libreOfficeCmd = `soffice --headless --convert-to pdf --outdir "${outputDir}" "${inputDocPath}"`;
        
        exec(libreOfficeCmd, (err) => {
            if (err) return reject(new Error('LibreOffice conversion failed: ' + err.message));
            
            const baseName = path.basename(inputDocPath, path.extname(inputDocPath));
            const pdfPath = path.join(outputDir, `${baseName}.pdf`);
            
            if (!fs.existsSync(pdfPath)) {
                return reject(new Error('Intermediate PDF file was not created.'));
            }
            
            // Step 2: Convert intermediate PDF pages to PNGs using pdftoppm
            const outputPrefix = path.join(outputDir, `${baseName}-page`);
            const pdftoppmCmd = `pdftoppm -png -r 150 "${pdfPath}" "${outputPrefix}"`;
            
            exec(pdftoppmCmd, (ppmErr) => {
                // Cleanup intermediate PDF
                fs.unlinkSync(pdfPath);
                
                if (ppmErr) return reject(new Error('pdftoppm rendering failed: ' + ppmErr.message));
                resolve(`Images successfully generated in: ${outputDir}`);
            });
        });
    });
}

<dependencies>
    <dependency>
        <groupId>org.apache.poi</groupId>
        <artifactId>poi-ooxml</artifactId>
        <version>5.2.3</version>
    </dependency>
    <dependency>
        <groupId>org.apache.poi</groupId>
        <artifactId>poi-scratchpad</artifactId>
        <version>5.2.3</version>
    </dependency>
</dependencies>

import org.apache.poi.xslf.usermodel.XMLSlideShow;
import org.apache.poi.xslf.usermodel.XSLFSlide;

import javax.imageio.ImageIO;
import java.awt.*;
import java.awt.geom.Rectangle2D;
import java.awt.image.BufferedImage;
import java.io.File;
import java.io.FileInputStream;
import java.io.FileOutputStream;
import java.io.IOException;
import java.util.List;

public class SlideToImageConverter {
    public static void convertSlidesToImages(File pptxFile, File outputFolder) throws IOException {
        try (FileInputStream is = new FileInputStream(pptxFile);
             XMLSlideShow ppt = new XMLSlideShow(is)) {
            
            Dimension pageSize = ppt.getPageSize();
            List<XSLFSlide> slides = ppt.getSlides();
            
            for (int i = 0; i < slides.size(); i++) {
                BufferedImage img = new BufferedImage(pageSize.width, pageSize.height, BufferedImage.TYPE_INT_ARGB);
                Graphics2D graphics = img.createGraphics();
                
                // Set default rendering options for high quality
                graphics.setRenderingHint(RenderingHints.KEY_ANTIALIASING, RenderingHints.VALUE_ANTIALIAS_ON);
                graphics.setRenderingHint(RenderingHints.KEY_RENDERING, RenderingHints.VALUE_RENDER_QUALITY);
                
                // Clear the canvas area
                graphics.setPaint(Color.white);
                graphics.fill(new Rectangle2D.Float(0, 0, pageSize.width, pageSize.height));
                
                // Draw the slide elements onto our graphic buffer
                slides.get(i).draw(graphics);
                
                File outputFile = new File(outputFolder, "slide-" + (i + 1) + ".png");
                try (FileOutputStream out = new FileOutputStream(outputFile)) {
                    ImageIO.write(img, "PNG", out);
                }
                graphics.dispose();
            }
        }
    }
}

sudo apt-get install -y pandoc weasyprint

import subprocess
import os

def render_docx_to_png(docx_path, output_png_path):
    temp_html = "temp_output.html"
    
    try:
        # Step 1: Convert DOCX to clean HTML via Pandoc
        subprocess.run([
            'pandoc', docx_path, '-f', 'docx', '-t', 'html', '-s', '-o', temp_html
        ], check=True)
        
        # Step 2: Compile the HTML structure to an image with WeasyPrint
        subprocess.run([
            'weasyprint', temp_html, output_png_path
        ], check=True)
        
    finally:
        if os.path.exists(temp_html):
            os.remove(temp_html)

pip install pdf2image

import subprocess
import os
from pdf2image import convert_from_path

def docx_to_images(docx_path, output_dir):
    # Convert DOCX to PDF using headless LibreOffice
    subprocess.run([
        'libreoffice', '--headless', '--convert-to', 'pdf', '--outdir', output_dir, docx_path
    ], check=True)
    
    # Track down the generated PDF file
    base_filename = os.path.splitext(os.path.basename(docx_path))[0]
    pdf_path = os.path.join(output_dir, f"{base_filename}.pdf")
    
    # Convert PDF pages to a list of PIL Images
    images = convert_from_path(pdf_path, dpi=150)
    
    generated_paths = []
    for i, image in enumerate(images):
        page_path = os.path.join(output_dir, f"{base_filename}_page_{i+1}.png")
        image.save(page_path, 'PNG')
        generated_paths.append(page_path)
        
    # Clean up intermediate PDF
    if os.path.exists(pdf_path):
        os.remove(pdf_path)
        
    return generated_paths