I have been trying to segment a docx document to multiple documents based on a predefined criteria. following is my approach to cut it to paragraphs
try {
FileInputStream in = new FileInputStream(file);
XWPFDocument doc = new XWPFDocument(in);
List<XWPFParagraph> paragraphs = doc.getParagraphs();
for (int idx = 0; idx < paragraphs.size(); idx++) {
XWPFDocument outputDocument = new XWPFDocument();
createParagraphInAnotherDocument(outputDocument, paragraphs.get(idx).getText());
String fullPath = String.format("./content/output/%1$s_%2$s_%3$04d.docx", FileUtils.getFileName(file), getName(), idx);
FileOutputStream outputStream = new FileOutputStream(fullPath);
outputDocument.write(outputStream);
outputDocument.close();
doc.close();
}
} catch (IOException e) {
e.printStackTrace();
}
While I am able to extract paragraphs with the code above, I can't find a way to extract pages. My understanding is that pages in word are render concern, and it happens in the runtime in the word application.
As far as I can see, the only way to do this is by interrogating the DOM model for the Word doc, and then determining how many paragraphs there are on each page. Below is a possible solution to the problem (it only works if the pages are explicitly separated by page breaks)
public static void main(String[] args) {
XWPFDocument doc = null;
try {
//Input Word Document
File file = new File("C:/TestDoc.docx");
FileInputStream in = new FileInputStream(file);
doc = new XWPFDocument(in);
//Determine how many paragraphs per page
List<Integer> paragraphCountList = getParagraphCountPerPage(doc);
if (paragraphCountList != null && paragraphCountList.size() > 0) {
int docCount = 0;
int startIndex = 0;
int endIndex = paragraphCountList.get(0);
//Loop through the paragraph counts for each page
for (int i=0; i < paragraphCountList.size(); i++) {
XWPFDocument outputDocument = new XWPFDocument();
List<XWPFParagraph> paragraphs = doc.getParagraphs();
List<XWPFParagraph> pageParagraphs = new ArrayList<XWPFParagraph>();
if (paragraphs != null && paragraphs.size() > 0) {
//Get the paragraphs from the input Word document
for (int j=startIndex; j < endIndex; j++) {
if (paragraphs.get(j) != null) {
pageParagraphs.add(paragraphs.get(j));
}
}
//Set the start and end point for the next set of paragraphs
startIndex = endIndex;
if (i < paragraphCountList.size()-2) {
endIndex = endIndex + paragraphCountList.get(i+1);
} else {
endIndex = paragraphs.size()-1;
}
//Create a new Word Doc with the paragraph subset
createPageInAnotherDocument(outputDocument, pageParagraphs);
//Write the file
String outputPath = "C:/TestDocOutput"+docCount+".docx";
FileOutputStream outputStream = new FileOutputStream(outputPath);
outputDocument.write(outputStream);
outputDocument.close();
docCount++;
pageParagraphs = new ArrayList<XWPFParagraph>();
}
}
}
} catch (Exception e) {
e.printStackTrace();
} finally {
try {
doc.close();
} catch (IOException ioe) {
ioe.printStackTrace();
}
}
}
private static List<Integer> getParagraphCountPerPage(XWPFDocument doc) throws Exception {
List<Integer> paragraphCountList = new ArrayList<>();
int paragraphCount = 0;
Document domDoc = convertStringToDOM(doc.getDocument().getBody().toString());
NodeList rootChildNodeList = domDoc.getChildNodes().item(0).getChildNodes();
for (int i=0; i < rootChildNodeList.getLength(); i++) {
Node childNode = rootChildNodeList.item(i);
if (childNode.getNodeName().equals("w:p")) {
paragraphCount++;
if (childNode.getChildNodes() != null) {
for (int k=0; k < childNode.getChildNodes().getLength(); k++) {
if (childNode.getChildNodes().item(k).getNodeName().equals("w:r")) {
for (int m=0; m < childNode.getChildNodes().item(k).getChildNodes().getLength(); m++) {
if (childNode.getChildNodes().item(k).getChildNodes().item(m).getNodeName().equals("w:br")) {
paragraphCountList.add(paragraphCount);
paragraphCount = 0;
}
}
}
}
}
}
}
paragraphCountList.add(paragraphCount+1);
return paragraphCountList;
}
private static Document convertStringToDOM(String xmlData) throws Exception {
DocumentBuilderFactory factory = DocumentBuilderFactory.newInstance();
DocumentBuilder builder = factory.newDocumentBuilder();
Document document = builder.parse(new InputSource(new StringReader(xmlData)));
return document;
}
private static void createPageInAnotherDocument(XWPFDocument outputDocument, List<XWPFParagraph> pageParagraphs) throws IOException {
for (int i = 0; i < pageParagraphs.size(); i++) {
addParagraphToDocument(outputDocument, pageParagraphs.get(i).getText());
}
}
private static void addParagraphToDocument(XWPFDocument outputDocument, String text) throws IOException {
XWPFParagraph paragraph = outputDocument.createParagraph();
XWPFRun run = paragraph.createRun();
run.setText(text);
}
来源:https://stackoverflow.com/questions/38047893/apache-poi-split-word-document-docx-to-pages