MultipartFile 实现word、excel转pdf

package com.pig4cloud.pigx.admin.utils;


import com.itextpdf.kernel.font.PdfFont;
import com.itextpdf.kernel.font.PdfFontFactory;
import com.itextpdf.kernel.geom.PageSize;
import com.itextpdf.kernel.pdf.PdfDocument;
import com.itextpdf.kernel.pdf.PdfWriter;
import com.itextpdf.layout.Document;
import com.itextpdf.layout.element.Cell;
import com.itextpdf.layout.element.Paragraph;
import com.itextpdf.layout.element.Table;
import org.apache.poi.hssf.usermodel.HSSFWorkbook;
import org.apache.poi.ss.usermodel.CellType;
import org.apache.poi.ss.usermodel.Row;
import org.apache.poi.ss.usermodel.Sheet;
import org.apache.poi.ss.usermodel.Workbook;
import org.apache.poi.xssf.usermodel.XSSFWorkbook;
import org.apache.poi.xwpf.usermodel.XWPFDocument;
import org.apache.poi.xwpf.usermodel.XWPFParagraph;
import org.springframework.mock.web.MockMultipartFile;
import org.springframework.web.multipart.MultipartFile;

import java.io.ByteArrayOutputStream;
import java.io.IOException;
import java.io.InputStream;
import java.util.Arrays;
import java.util.List;
import java.util.Objects;

/**
 * @author xiechenyu
 * @Description：
 * @date 2024/5/22 9:27
 */
public class ConvertPdfUtil {
	private static final List<String> allowedFileExtensions = Arrays.asList("xls", "xlsx", "doc", "docx");

	public static MultipartFile convertToPdf(MultipartFile file) {
		String originalFilename = file.getOriginalFilename();
		if (Objects.isNull(originalFilename)) {
			throw new RuntimeException("请上传正确文件");
		}
		checkFileExtensions(originalFilename);

		// 转换为PDF文件
		String fileExtension = getFileExtension(originalFilename);
		if (Objects.equals(fileExtension, allowedFileExtensions.get(0)) || Objects.equals(fileExtension, allowedFileExtensions.get(1))) {
			return excelToPdf(file);
		} else if (Objects.equals(fileExtension, allowedFileExtensions.get(2)) || Objects.equals(fileExtension, allowedFileExtensions.get(3))) {
			return wordToPdf(file);
		}
		return file;
	}

	//word文件转pdf，支持doc，docx
	private static MultipartFile wordToPdf(MultipartFile file) {
		try (ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream();
			 PdfDocument pdf = new PdfDocument(new PdfWriter(byteArrayOutputStream));
			 Document pdfDocument = new Document(pdf, PageSize.A4.rotate());
			 InputStream in = file.getInputStream();
			 XWPFDocument word = new XWPFDocument(in);) {
			//指定pdf的字体样式为STSong-Light，编码方式为UniGB-UCS2-H,不指定编码方式中文乱码
			PdfFont pdfFont = PdfFontFactory.createFont("STSong-Light", "UniGB-UCS2-H");

			// 逐段写入PDF
			for (XWPFParagraph paragraph : word.getParagraphs()) {
				pdfDocument.add(new Paragraph(paragraph.getText()).setFont(pdfFont));
			}
			pdfDocument.close();

			String fileExtension = getFileExtension(Objects.requireNonNull(file.getOriginalFilename()));
			byte[] abbreviateByteArray = byteArrayOutputStream.toByteArray();
			return new MockMultipartFile(file.getName(), file.getOriginalFilename().replace("" + fileExtension, "pdf"), "application/pdf", abbreviateByteArray);
		} catch (IOException e) {
			throw new RuntimeException(e);
		}
	}


	//excel文件转pdf，支持xsl，xslx
	private static MultipartFile excelToPdf(MultipartFile file) {
		try (ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream();
			 PdfDocument pdf = new PdfDocument(new PdfWriter(byteArrayOutputStream));
			 Document document = new Document(pdf, PageSize.A4.rotate());
			 InputStream in = file.getInputStream();) {

			//指定pdf的字体样式为STSong-Light，编码方式为UniGB-UCS2-H
			PdfFont pdfFont = PdfFontFactory.createFont("STSong-Light", "UniGB-UCS2-H");

			//excel文件类型有xls和xlsx两种，对应的构造器不一样
			String fileExtension = getFileExtension(Objects.requireNonNull(file.getOriginalFilename()));
			Workbook workbook = Objects.equals(fileExtension, ".xsl") ? new HSSFWorkbook(in) : new XSSFWorkbook(in);

			int numberOfSheets = workbook.getNumberOfSheets();
			for (int k = 0; k < numberOfSheets; k++) {
				Sheet sheet = workbook.getSheetAt(k);
				//获取总列数
				int column = getMaxColumn(sheet);
				//创建一个pdf的table对象，以便在pdf中以表的形式展示，列数和excel列数一致
				Table table = new Table(column);

				for (int i = sheet.getFirstRowNum(); i <= sheet.getLastRowNum(); i++) {
					for (int j = 0; j < column; j++) {
						String str = "";
						//获取excel单元格
						Row row = sheet.getRow(i);
						if (Objects.isNull(row)) {
							continue;
						}
						org.apache.poi.ss.usermodel.Cell cell = row.getCell(j);
						if (Objects.nonNull(cell)) {
							CellType cellType = cell.getCellType();
							if (cellType == CellType.NUMERIC) {
								str = (int) cell.getNumericCellValue() + "";
							} else if (cellType == CellType.FORMULA) {
								switch (cell.getCachedFormulaResultType()) {
									case NUMERIC:
										str = String.valueOf(cell.getNumericCellValue());
										break;
									case STRING:
										str = cell.getRichStringCellValue().getString();
										break;
									default:
										str = "";
										break;
								}
							} else {
								str = cell.getStringCellValue();
							}
						}
						//将pdf的单元格初始化并添加到pdf中
						Cell cells = new Cell().setFont(pdfFont).add(new Paragraph(str));
						table.addCell(cells);
					}
				}
				//两个table之间增加一个空格
				document.add(new Paragraph(" "));
				document.add(table);
			}

			document.close();
			byte[] abbreviateByteArray = byteArrayOutputStream.toByteArray();
			//这里的contentType必须为application/pdf，不然会格式损坏
			return new MockMultipartFile(file.getName(), file.getOriginalFilename().replace("." + fileExtension, ".pdf"), "application/pdf", abbreviateByteArray);

		} catch (IOException e) {
			throw new RuntimeException(e);
		}
	}

	private static int getMaxColumn(Sheet sheet) {
		int maxColumn = 0;
		for (int i = sheet.getFirstRowNum(); i <= sheet.getLastRowNum(); i++) {
			if (Objects.nonNull(sheet.getRow(i))) {
				maxColumn = Math.max(maxColumn, sheet.getRow(i).getLastCellNum());
			}
		}
		return maxColumn;
	}


	private static void checkFileExtensions(String originalFilename) {
		String fileExtension = getFileExtension(originalFilename);
		if (!allowedFileExtensions.contains(fileExtension)) {
			throw new RuntimeException("文件格式不正确,支持转换为PDF文件的格式有：xls、xlsx、doc、docx");
		}

	}

	private static String getFileExtension(String filename) {
		if (filename.lastIndexOf('.') != -1 && filename.lastIndexOf('.') != 0) {
			return filename.substring(filename.lastIndexOf('.') + 1);
		} else {
			return "";
		}
	}


}