YunNanDsBase/src/main/java/com/dsideal/base/Test/ReadDoc.java

package com.dsideal.base.Test;

import org.apache.poi.openxml4j.exceptions.InvalidFormatException;
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 org.apache.poi.xwpf.usermodel.XWPFChart;
import org.apache.poi.xwpf.usermodel.XWPFDocument;
import org.apache.poi.xwpf.usermodel.XWPFParagraph;

import java.io.*;
import java.util.List;

public class ReadDoc {
    //https://blog.csdn.net/a346736962/article/details/123037797
    public static void main(String[] args) throws IOException, InvalidFormatException {
        String workingPath = "D:\\dsWork\\YunNanDsBase\\Doc\\全省及州市县区人口与教育报告集20241023\\16个州市报告2022\\分析报告20240510";
        //遍历workingPath下的所有文件
        File file = new File(workingPath);
        String[] fileList = file.list();
        for (String fileName : fileList) {
            //判断是否为文件夹
            File tempFile = new File(workingPath + "/" + fileName);
            if (tempFile.isDirectory()) {
                continue;
            }
            //判断是否为docx文件
            if (fileName.endsWith(".docx") && !fileName.startsWith("~")) {
                //读取文件
                String inputUrl = workingPath + "/" + fileName;
                InputStream is = new FileInputStream(inputUrl);
                XWPFDocument doc = new XWPFDocument(is);
                //图表
                List<XWPFChart> charts = doc.getCharts();

                System.out.println("图表数量="+charts.size());

                for (XWPFChart chart : charts) {
                    XSSFWorkbook workbook = chart.getWorkbook();
                    XSSFSheet sheet = workbook.getSheetAt(0);
                    //遍历一下sheet
                    for (int i = 0; i < sheet.getPhysicalNumberOfRows(); i++) {
                        //遍历行
                        XSSFRow row = sheet.getRow(i);
                        if(row==null) continue;
                        for (int j = 0; j < row.getPhysicalNumberOfCells(); j++) {
                            //遍历列
                            XSSFCell cell = row.getCell(j);
                            if (cell != null)
                                System.out.print(cell + " ");
                            else{
                                System.out.print("null ");
                            }
                        }
                        System.out.println();
                    }
                }
                //段落
                List<XWPFParagraph> paragraphs = doc.getParagraphs();
                for (XWPFParagraph paragraph : paragraphs) {
                    String text = paragraph.getText();
                    if(text.startsWith("图")){
                        System.out.println(text);
                    }

                }
            }
        }

    }
}