读取PDF中的excel表格，并写入excel中

import com.spire.pdf.FileFormat; import com.spire.pdf.PdfDocument; import com.spire.pdf.utilities.PdfTable; import com.spire.pdf.utilities.PdfTableExtractor; //import com.spire.xls.ExcelVersion; //import com.spire.xls.*; import org.apache.commons.io.FileUtils; import org.apache.poi.xssf.usermodel.XSSFCell; import org.apache.poi.xssf.usermodel.XSSFRow; import org.apache.poi.xssf.usermodel.XSSFSheet; import org.apache.poi.xssf.usermodel.XSSFWorkbook; import java.io.*; /** * 1.读取pdf里面的excel表格 * 2.写入excel中 * 3.读取excel数据 */ public class ExcelCompare { public static void main(String[] args) { String filePath = "E:\\tmp\\test1.pdf"; String textPath = "E:\\tmp\\test1.txt"; String excelPath = "E:\\tmp\\test1.xlsx"; String excelPath2 = "E:\\tmp\\test2.xlsx"; // // 加载pdf文件 // PdfDocument pdf = new PdfDocument(); // pdf.loadFromFile(filePath); // //保存为Excel文档 // pdf.saveToFile(excelPath, FileFormat.XLSX); // pdf.dispose(); //加载PDF文档 PdfDocument pdf = new PdfDocument(); pdf.loadFromFile(filePath); //创建StringBuilder类的实例 StringBuilder builder = new StringBuilder(); //抽取表格 PdfTableExtractor extractor = new PdfTableExtractor(pdf); PdfTable[] tableLists ; // PDF总页数 int count = pdf.getPages().getCount(); if(count > 0) { // 创建一个工作簿对象 // Workbook workbook = new Workbook(); XSSFWorkbook workbook = new XSSFWorkbook(); // 删除默认工作表 // workbook..getWorksheets().clear(); for (int page = 0; page < count; page++) { tableLists = extractor.extractTable(page); if (tableLists != null && tableLists.length > 0) { for (PdfTable table : tableLists) { // 创建EXCEL的sheet页 String sheetName = String.format("sheetName - %d", page); // Worksheet sheet = workbook.getWorksheets().add(sheetName); XSSFSheet sheet = workbook.createSheet(); int row = table.getRowCount(); int column = table.getColumnCount(); for (int i = 0; i < row; i++) { // 创建行 XSSFRow head = sheet.createRow(i); XSSFCell cell = null; for (int j = 0; j < column; j++) { String text = table.getText(i, j); builder.append(text+" "); //将数据插入特定单元格 // sheet.get(i + 1, j + 1).setText(text); cell = head.createCell(j); cell.setCellValue(text); } builder.append("\r\n"); } //自动调整列宽 // for (int sheetColNum = 0; sheetColNum < sheet.getColumns().length; sheetColNum++) { // sheet.autoFitColumn(sheetColNum + 1); // } for (int sheetColNum = 0; sheetColNum < sheet.getRow(1).getLastCellNum(); sheetColNum++) { sheet.autoSizeColumn((short)sheetColNum, true); } } } } // 保存前判断是否已经存在EXCEL文件，存在则删除 File fileOne = new File(excelPath2); if (fileOne.exists() && fileOne.isFile()) { fileOne.delete(); } // 将工作簿保存为 Excel 文件 // workbook.saveToFile(excelPath2, ExcelVersion.Version2016); try { fileOne.createNewFile(); //文件输出流 FileOutputStream stream = FileUtils.openOutputStream(fileOne); //写入 workbook.write(stream); //关闭输出流 stream.close(); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } System.out.println("将" + filePath + "转换为EXCEL结束！！！！！"); } //将提取的表格内容写入txt文档 FileWriter fileWriter = null; try { fileWriter = new FileWriter(textPath); fileWriter.write(builder.toString()); fileWriter.flush(); fileWriter.close(); } catch (IOException e) { e.printStackTrace(); } // 读取excel FileInputStream xlsStream = null; try { // Excel工作簿 输入流 xlsStream = new FileInputStream(new File("E:\\tmp\\test_tmp.xlsx")); // 构造工作簿对象 XSSFWorkbook xssfWorkbook = new XSSFWorkbook(xlsStream); // 获取工作表,这里获取的是第一个sheet， // 如果一个工作薄对象有多个sheet的话，就需要遍历获取多个 XSSFSheet sheetAt = xssfWorkbook.getSheetAt(0); // 获取行,行号作为参数传递给getRow方法 XSSFRow row = sheetAt.getRow(0); // 获取单元格,row已经确定了行号,列号作为参数传递给getCell，就可以获得相应的单元格了 XSSFCell readCell = row.getCell(0); // 获取单元格的值 String cellValue = readCell.getStringCellValue(); System.out.println("获取到的数据是：" + cellValue); //遍历excel for (int i = 0; i <= sheetAt.getLastRowNum(); i++) { row = sheetAt.getRow(i); for (int j = 0; j < row.getLastCellNum(); j++) { readCell = row.getCell(j); System.out.println("第" + (i+1) + "行，第"+ j + "列，值为：" + readCell); } } } catch (FileNotFoundException e) { // TODO Auto-generated catch block e.printStackTrace(); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); }finally { if (xlsStream != null) { try { xlsStream.close(); } catch (IOException e) { // TODO Auto-generated catch block e.printStackTrace(); } } } } }