java 提取 word 章节内容_java读取word文档，提取标题和内容

packagecom.w.test;importjava.io.File;importjava.io.FileInputStream;importjava.io.InputStream;importjava.util.ArrayList;importjava.util.List;importjava.util.regex.Matcher;importjava.util.regex.Pattern;importorg.apache.poi.POIXMLDocument;importorg.apache.poi.POIXMLTextExtractor;importorg.apache.poi.hwpf.HWPFDocument;importorg.apache.poi.hwpf.extractor.WordExtractor;importorg.apache.poi.hwpf.usermodel.CharacterRun;importorg.apache.poi.hwpf.usermodel.Paragraph;importorg.apache.poi.hwpf.usermodel.Range;importorg.apache.poi.openxml4j.opc.OPCPackage;importorg.apache.poi.poifs.filesystem.POIFSFileSystem;importorg.apache.poi.xwpf.extractor.XWPFWordExtractor;importorg.apache.poi.xwpf.usermodel.XWPFDocument;importorg.apache.poi.xwpf.usermodel.XWPFParagraph;importorg.apache.poi.xwpf.usermodel.XWPFRun;importcom.example.model.Policy_content;public classGetWord {public static voidmain(String[] args) {//TODO Auto-generated method stub

try{

List list = new ArrayList<>();

InputStream is= new FileInputStream(new File("文件路径")); //需要将文件路更改为word文档所在路径。

POIFSFileSystem fs= newPOIFSFileSystem(is);

HWPFDocument document= newHWPFDocument(fs);

Range range=document.getRange();

CharacterRun run1= null;//用来存储第一行内容的属性

CharacterRun run2 = null;//用来存储第二行内容的属性

int q=1;for (int i = 0; i < range.numParagraphs()-1; i++) {

Paragraph para1= range.getParagraph(i);//获取第i段

Paragraph para2 = range.getParagraph(i+1);//获取第i段

int t=i; //记录当前分析的段落数

String paratext1= para1.text().trim().replaceAll("\r\n", ""); //当前段落和下一段

String paratext2 = para2.text().trim().replaceAll("\r\n", "");

run1=para1.getCharacterRun(0);

run2=para2.getCharacterRun(0);if (paratext1.length() > 0&&paratext2.length() > 0) {//这个if语句为的是去除大标题，连续三个段落字体大小递减就跳过

if(run1.getFontSize()>run2.getFontSize()&&run2.getFontSize()>range.getParagraph(i+2).getCharacterRun(0).getFontSize()) {continue;

}//连续两段字体格式不同

if(run1.getFontSize()>run2.getFontSize()) {

String content=paratext2;

run1=run2; //从新定位run1 run2

run2=range.getParagraph(t+2).getCharacterRun(0);

t=t+1;while(run1.getFontSize()==run2.getFontSize()) {//连续的相同

content+=range.getParagraph(t+1).text().trim().replaceAll("\r\n", "");

run1=run2;

run2=range.getParagraph(t+2).getCharacterRun(0);

t++;

}if(paratext1.indexOf("HYPERLINK")==-1&&content.indexOf("HYPERLINK")==-1) {

System.out.println(q+"标题"+paratext1+"\t内容"+content);

i=t;

q++;

}

}

}

}

}catch(Exception e) {

e.printStackTrace();

}

}

}