update 更改模块名称

2023-09-05 09:10:12 +08:00
parent 248a2dbe8a
commit 8522fc6f34
1501 changed files with 4 additions and 47397 deletions
@@ -0,0 +1,37 @@
+package com.jero.modules.document.utils;
+
+import com.jero.common.exception.JeroBootException;
+import com.jero.common.util.MinioUtil;
+import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.text.PDFTextStripper;
+import org.apache.pdfbox.text.PDFTextStripperByArea;
+
+import java.io.IOException;
+import java.io.InputStream;
+
+/**
+ * @description
+ * @date 2022/3/21 17:04
+ * @auth zhn
+ */
+public class ReadPdfUtil {
+
+    public static String readPdf(String path) {
+        if(MinioUtil.doesObjectExist(path)){
+            try (InputStream in = MinioUtil.download(path);
+                 PDDocument document = PDDocument.load(in)) {
+                document.getClass();
+                PDFTextStripperByArea stripper = new PDFTextStripperByArea();
+                stripper.setSortByPosition(true);
+                PDFTextStripper tStripper = new PDFTextStripper();
+
+                String pdfFileInText = tStripper.getText(document);
+                return pdfFileInText;
+
+            }catch (IOException e) {
+                throw new JeroBootException("文件内容读取失败");
+            }
+        }
+        return null;
+    }
+}
@@ -0,0 +1,96 @@
+package com.jero.modules.document.utils;
+
+import com.jero.common.exception.JeroBootException;
+import com.jero.common.util.MinioUtil;
+import org.apache.poi.hwpf.extractor.WordExtractor;
+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.InputStream;
+
+import static com.jero.modules.document.service.impl.BussDocumentLibraryEOServiceImpl.copyFile;
+
+/**
+ * @description
+ * @date 2022/3/21 17:04
+ * @auth zhn
+ */
+public class ReadWordUtil {
+    public static String readWord(String path,String uploadpath) {
+//        String path = "E:\\DeskTop\\1111.doc";
+//        path = "E:\\DeskTop\\22222.docx";
+//        String path = "E:\\DeskTop\\标准所ISO-用户手册.doc";
+        String str = "";
+        try {
+            if (path.endsWith(".doc")) {
+                if(MinioUtil.doesObjectExist(path)){
+                    InputStream in = MinioUtil.download(path);
+                    WordExtractor ex = new WordExtractor(in);
+                    str = ex.getText();
+                    ex.close();
+                    in.close();
+                }
+            } else if (path.endsWith("docx")) {
+                //先把文件存到本地, 用完后在删除
+                if(MinioUtil.doesObjectExist(path)){
+                    InputStream in = MinioUtil.download(path);
+                    String fileNowPath = uploadpath + "/tempZip/" ;
+                    File file = new File(fileNowPath);
+                    if (file.exists()) {
+                        file.delete();
+                    }
+                    file.mkdirs();
+                    File fileOld = new File(path);
+                    long currentTime = System.currentTimeMillis();
+                    copyFile(in, fileNowPath + File.separator + currentTime + fileOld.getName());
+                    File fileTemp = new File(fileNowPath + File.separator + currentTime + fileOld.getName());
+                    String pathTemp = fileTemp.getPath();
+                    OPCPackage opcPackage = POIXMLDocument.openPackage(pathTemp);
+                    POIXMLTextExtractor extractor = new XWPFWordExtractor(opcPackage);
+                    str = extractor.getText().trim();
+                    extractor.close();
+                    fileTemp.delete();
+                    in.close();
+                }
+            } else {
+                throw new JeroBootException("此文件不是word文件");
+            }
+            return str;
+        } catch (Exception e) {
+            e.printStackTrace();
+            throw new JeroBootException("文件内容读取失败");
+        }catch (NoSuchFieldError e){
+            e.printStackTrace();
+            throw new JeroBootException("文件内容读取失败");
+        }
+    }
+//    public static String readWord(String path) {
+////        String path = "E:\\DeskTop\\1111.doc";
+////        path = "E:\\DeskTop\\22222.docx";
+////        String path = "E:\\DeskTop\\标准所ISO-用户手册.doc";
+//        String str = "";
+//        try {
+//            if (path.endsWith(".doc")) {
+//                InputStream is = new FileInputStream(new File(path));
+//                WordExtractor ex = new WordExtractor(is);
+//                str = ex.getText();
+//                ex.close();
+//            } else if (path.endsWith("docx")) {
+//                OPCPackage opcPackage = POIXMLDocument.openPackage(path);
+//                POIXMLTextExtractor extractor = new XWPFWordExtractor(opcPackage);
+//                str = extractor.getText();
+//                extractor.close();
+//            } else {
+//                throw new JeroBootException("此文件不是word文件");
+//            }
+//            return str;
+//        } catch (Exception e) {
+//            throw new JeroBootException("文件内容读取失败");
+//        }catch (NoSuchFieldError e){
+//            throw new JeroBootException("文件内容读取失败");
+//        }
+//    }
+}