poi解析word

最新推荐文章于 2024-06-06 09:00:00 发布

我是小何啊

最新推荐文章于 2024-06-06 09:00:00 发布

阅读量1.5k

点赞数

分类专栏：笔记文章标签： apache java 开发语言

本文链接：https://blog.csdn.net/qq_46036369/article/details/121560423

版权

poi解析word

- 读取文件所有内容
- 读取文件中所有表格及其内容

读取文件所有内容

import org.apache.poi.ooxml.POIXMLDocument;
import org.apache.poi.ooxml.extractor.POIXMLTextExtractor;
import org.apache.poi.openxml4j.opc.OPCPackage;
import org.apache.poi.xwpf.extractor.XWPFWordExtractor;
import java.io.File;
import java.io.FileInputStream;
import java.io.IOException;

/**
 * @author: ThinkPad
 * @date: 2021/11/19 11:27
 * @description:
 */
public class WordTest {
   
    /**
     * 读取doc文件内容
     *
     * @param
     * @return 返回文件内容
     * @throws IOException
     */
    public static String doc2String(FileInputStream fs) throws IOException {
   
        String path = "C:\\Users\\Administrator\\Desktop\\11-17\\NA7029 Order Details.docx";
        OPCPackage opcPackage = POIXMLDocument.openPackage(path);
        POIXMLTextExtractor extractor = new XWPFWordExtractor(opcPackage);
        String buffer = extractor.getText();
        System.out.println(buffer);
        extractor.close();
        return null;
    }
    public static String doc2String(File file) throws IOException {
   
        return doc2String(new FileInputStream(file));
    }

    public static void main(String[] args) {
   
        File file = new File("C:\\Users\\Administrator\\Desktop\\11-17\\NA7029 Order Details.docx");
        try {
   
            System.out.println(doc2String(file));
        } catch (IOException e) {
   
            e.printStackTrace();
        }
    }

}

读取文件中所有表格及其内容

package com.chenmei.util;

import com.chenmei.bean.ExcelAll;
import com.chenmei.bean.ExcelNotes;
import com.chenmei.bean.ExeclOrder;
import com.chenmei.bean.ExeclTop;
import org.apache.poi.ooxml.POIXMLDocument;
import org.apache.poi.ooxml.extractor.POIXMLTextExtractor;
import org.apache.poi.openxml4j.opc.OPCPackage;
import org.apache.poi.xwpf.extractor.XWPFWordExtractor;
import org.apache.poi.xwpf.usermodel.XWPFDocument;
import org.apache.poi.xwpf.usermodel.XWPFTable;
import org.apache.poi.xwpf.usermodel.XWPFTableCell;
import org.apache.poi.xwpf.usermodel.XWPFTableRow;

import java.io.FileInputStream;
import java.util.*;

import static java.util.stream.Collectors.toMap;

/**
 * @author: ThinkPad
 * @date: 2021/9/26 14:09
 * @description:
 */
public class WordUtil {
   
    public ExcelAll getWord(String filePath){
   
    HashMap<String, String> materialMap = new HashMap<String, String>() {
   {
   
            put("Pine", "松木");
            put("Basswood", "椴木");
            put("Poly", "PVC");
            put("Waterproof Poly", "PVC");
            put("White Ash", "水曲柳");
            put("Abachi", "阿尤斯");
            put("Paulownia", "桐木");
        }};
        HashMap<String, String> installationStyleMap = new HashMap<String, String>() {
   {
   
            put("Full height", "Full height(全部高度)");
            put("Café Style", "Café Style");
            put("T on T without T-Post", "T on T without T-Post (双层窗没有加T型柱)");
            put("By pass track system", "By pass track system (推拉窗）");
            put("Bi Fold track system", "Bi Fold track system(折叠窗)");
            put("Shaped shutters", "Shaped shutters(异形窗)");
        }};
        HashMap<String, String> inOrOutOfRecessMap = new HashMap<String, String>() {
   {
   
            put("Inside", "Inside(内装)");
            put("Outside", "Outside(外装)");
        }};
        HashMap<String, String> colourMap = new HashMap<String, String>() {
   {
   
            put(