package com.dsideal.base.Tools.Test; import com.dsideal.base.Tools.FillData.ExcelKit.ExcelKit; import com.jfinal.kit.StrKit; import org.apache.poi.openxml4j.exceptions.InvalidFormatException; import org.dom4j.Document; import org.dom4j.DocumentException; import org.dom4j.Element; import org.dom4j.io.SAXReader; import org.xml.sax.SAXException; import javax.xml.parsers.ParserConfigurationException; import javax.xml.xpath.XPathExpressionException; import java.io.File; import java.io.FileInputStream; import java.io.FileOutputStream; import java.io.IOException; import java.util.ArrayList; import java.util.List; import java.util.zip.ZipEntry; import java.util.zip.ZipInputStream; public class TestOutSideExcel { /** * 解压缩 * * @throws IOException */ public static void UnCompress(String wordPath, String workingPath) throws IOException { workingPath = workingPath.replace("\\", "/"); if (!workingPath.endsWith("/")) workingPath += "/"; File file = new File(wordPath);//取得word文件 FileInputStream inputStream = new FileInputStream(file); ZipInputStream zipInputStream = new ZipInputStream(inputStream); ZipEntry entry; byte[] ch = new byte[256]; while ((entry = zipInputStream.getNextEntry()) != null) { File zFile = new File(workingPath + entry.getName()); if (entry.isDirectory()) { if (!zFile.exists()) { zFile.mkdirs(); } zipInputStream.closeEntry(); } else { File fpath = new File(zFile.getParent()); if (!fpath.exists()) { fpath.mkdirs(); } FileOutputStream outputStream = new FileOutputStream(zFile); int i; while ((i = zipInputStream.read(ch)) != -1) { outputStream.write(ch, 0, i); } zipInputStream.closeEntry(); outputStream.close(); } } inputStream.close(); } /** * 读取指定的Chart数据 * * @param workingPath * @param chartNumber */ public static List> readChar(String workingPath, int chartNumber) throws DocumentException { List> res = new ArrayList<>(); String xml = workingPath + "\\word\\charts\\chart" + chartNumber + ".xml"; if (!(new File(xml).exists())) { System.out.println("没有找到第" + chartNumber + "个图表"); return res; } //3、开始读取 // 创建 SAXReader 对象,读取 XML 文件 SAXReader reader = new SAXReader(); Document document = reader.read(new File(xml)); // 获取根元素 Element root = document.getRootElement(); //折线图 //将xml用IDEA打开,搜索关键的数据值,然后右键查看XPATH完整路径可以获取到下面的路径 ///c:chartSpace/c:chart/c:plotArea/c:lineChart/c:ser/c:cat/c:numRef/c:numCache/c:pt/c:v //在每个已知的图表类型中查找,找到后跳出循环 //声明一个数组,图表的所有类型 String[] CHART_TYPES = {"lineChart", "barChart"};//折线,柱状 String type = ""; for (String chartType : CHART_TYPES) { if (root.element("chart").element("plotArea") .element(chartType) != null) { type = chartType; break; } } if (StrKit.isBlank(type)) { System.out.println("没有找到图表类型,请扩充图表类型"); System.out.println(root.element("chart").element("plotArea")); System.exit(-1); } List xList = root.element("chart").element("plotArea") .element(type).element("ser").element("cat") .element("numRef").element("numCache").elements("pt"); List> allValueList = new ArrayList<>(); ///c:chartSpace/c:chart/c:plotArea/c:lineChart/c:ser/c:val/c:numRef/c:numCache for (Element ser : root.element("chart").element("plotArea") .element(type).elements("ser")) { List yList = ser.element("val") .element("numRef").element("numCache").elements("pt"); allValueList.add(yList); } for (int i = 0; i < xList.size(); i++) { List row = new ArrayList<>(); row.add(xList.get(i).element("v").getText()); for (int j = 0; j < allValueList.size(); j++) { List yList = allValueList.get(j); if (yList == null || i + 1 > yList.size() || yList.get(i) == null) { row.add(null); continue; } Element e = yList.get(i).element("v"); if (e != null) { row.add(e.getText()); } else { row.add(null); } } res.add(row); } return res; } public static void main(String[] args) throws IOException, InvalidFormatException, InterruptedException, ParserConfigurationException, SAXException, XPathExpressionException, DocumentException { String sourceDoc = "c:/西双版纳州人口变化及其对教育的影响20240420.docx"; //1、将word文件解压缩 String workingPath = "C:\\zipFile"; UnCompress(sourceDoc, workingPath); //2、我们需要第几个图表 for (int chartNumber = 4; chartNumber <= 4; chartNumber++) { System.out.println("正在处理第" + chartNumber + "个图表的信息~"); //读取图表 List> list = readChar(workingPath, chartNumber); ExcelKit.printTable(list); System.out.println("========================================================="); } } }