server_commit

src/main/java/io/v/nutz/base/utils/WordUtil.java

@@ -0,0 +1,99 @@
+package io.v.nutz.base.utils;
+
+import fr.opensagres.poi.xwpf.converter.xhtml.XHTMLConverter;
+import lombok.extern.slf4j.Slf4j;
+import org.apache.poi.hwpf.HWPFDocument;
+import org.apache.poi.hwpf.converter.WordToHtmlConverter;
+import org.apache.poi.xwpf.usermodel.XWPFDocument;
+import org.apache.poi.xwpf.usermodel.XWPFPictureData;
+import org.jsoup.Jsoup;
+import org.jsoup.nodes.Document;
+import org.jsoup.nodes.Element;
+import org.jsoup.select.Elements;
+
+import javax.xml.parsers.DocumentBuilderFactory;
+import javax.xml.transform.OutputKeys;
+import javax.xml.transform.Transformer;
+import javax.xml.transform.TransformerFactory;
+import javax.xml.transform.dom.DOMSource;
+import javax.xml.transform.stream.StreamResult;
+import java.io.ByteArrayOutputStream;
+import java.io.File;
+import java.io.FileInputStream;
+import java.nio.charset.StandardCharsets;
+import java.util.Base64;
+import java.util.List;
+
+/**
+ * @author jug
+ * @date 2024/02/29
+ */
+@Slf4j
+public class WordUtil {
+
+    public static String checkConvert2Html(File file) throws Exception {
+        String fileName = file.getName().toLowerCase();
+        if (fileName.endsWith(".doc")) {
+            return doc2Html(file);
+        } else if (fileName.endsWith(".docx")) {
+            return docx2Html(file);
+        } else {
+            return "";
+        }
+    }
+
+    public static String doc2Html(File file) throws Exception {
+        try (FileInputStream inputStream = new FileInputStream(file)) {
+            HWPFDocument wordDocument = new HWPFDocument(inputStream);
+            WordToHtmlConverter converter = new WordToHtmlConverter(DocumentBuilderFactory.newInstance().newDocumentBuilder().newDocument());
+            converter.setPicturesManager((bytes, pictureType, s, v, v1) -> {
+                String type = pictureType.name();
+                return "data:image/" + type + ";base64," + Base64.getEncoder().encodeToString(bytes);
+            });
+            converter.processDocument(wordDocument);
+            Transformer transformer = TransformerFactory.newInstance().newTransformer();
+            transformer.setOutputProperty(OutputKeys.ENCODING, "utf-8");
+            transformer.setOutputProperty(OutputKeys.INDENT, "yes");
+            transformer.setOutputProperty(OutputKeys.METHOD, "html");
+            ByteArrayOutputStream outputStream = new ByteArrayOutputStream();
+            transformer.transform(new DOMSource(converter.getDocument()), new StreamResult(outputStream));
+            return outputStream.toString(StandardCharsets.UTF_8);
+        } catch (Exception e) {
+            log.error("Error converting doc to html: " + e.getMessage());
+            throw e;
+        }
+    }
+
+    public static String docx2Html(File file) throws Exception {
+        try (FileInputStream inputStream = new FileInputStream(file);
+             XWPFDocument document = new XWPFDocument(inputStream)) {
+
+            List<XWPFPictureData> list = document.getAllPictures();
+            ByteArrayOutputStream outputStream = new ByteArrayOutputStream();
+            XHTMLConverter.getInstance().convert(document, outputStream, null);
+            String html = new String(outputStream.toByteArray());
+            Document doc = Jsoup.parse(html);
+            Elements elements = doc.getElementsByTag("img");
+
+            if (elements != null && elements.size() > 0 && list != null) {
+                for (Element element : elements) {
+                    String src = element.attr("src");
+                    for (XWPFPictureData data : list) {
+                        if (src.contains(data.getFileName())) {
+                            String type = src.substring(src.lastIndexOf(".") + 1);
+                            String base64 = "data:image/" + type + ";base64," + Base64.getEncoder().encodeToString(data.getData());
+                            element.attr("src", base64);
+                            break;
+                        }
+                    }
+                }
+            }
+            return doc.html();
+        } catch (Exception e) {
+            log.error("Error converting docx to html: " + e.getMessage());
+            throw e;
+        }
+    }
+
+
+}