编写所有基础工具，提供word util pdf util

src/main/java/com/docus/sw/fenpan/FenpanService.java

@@ -2,6 +2,7 @@ package com.docus.sw.fenpan;
 
 import com.alibaba.excel.util.FileUtils;
 import com.docus.sw.Config;
+import com.docus.sw.word.GetPicsDocx;
 import com.docus.sw.word.PdfBoxUtils;
 import lombok.extern.slf4j.Slf4j;
 import org.apache.commons.imaging.ImageInfo;
@@ -102,10 +103,52 @@ public class FenpanService {
 
         for (Pieces piece : pieces) {
             //根据文件类型
-            if (piece.getFileTypeEnum() == FileTypeEnum.WORD) {
+            if (piece.getFileTypeEnum() == FileTypeEnum.DOC) {
                 //从word 直接提取图片
                 //提取图片为document ，然后用于后面判断
+                File file = new File("temp");
+                if(!file.exists()){
+                    file.mkdirs();
+                }
+
+                GetPicsDocx.getPics(piece.getAbsolutePath(),file.getAbsolutePath());
+                List<Document> documentList = new ArrayList<>();
+                try {
+                    PdfBoxUtils.pdf2image(piece.getAbsolutePath(),file.getAbsolutePath());
+                    File[] files = file.listFiles();
+                    for(File pdfImg:files){
+                        getDocumentList(documentList, pdfImg);
+                    }
+                } catch (IOException e) {
+                    log.error(e.getMessage(),e);
+                }
+
+                piece.put(documentList);
+                //删除对应的temp 文件
+                file.delete();
+            } else  if (piece.getFileTypeEnum() == FileTypeEnum.DOCX) {
+                //从word 直接提取图片
+                //提取图片为document ，然后用于后面判断
+                File file = new File("temp");
+                if(!file.exists()){
+                    file.mkdirs();
+                }
+
+                GetPicsDocx.getPics(piece.getAbsolutePath(),file.getAbsolutePath());
+                List<Document> documentList = new ArrayList<>();
+                try {
+                    PdfBoxUtils.pdf2image(piece.getAbsolutePath(),file.getAbsolutePath());
+                    File[] files = file.listFiles();
+                    for(File pdfImg:files){
+                        getDocumentList(documentList, pdfImg);
+                    }
+                } catch (IOException e) {
+                    log.error(e.getMessage(),e);
+                }
 
+                piece.put(documentList);
+                //删除对应的temp 文件
+                file.delete();
             } else if (piece.getFileTypeEnum() == FileTypeEnum.PDF) {
                 // 从pdf 提取图片,
                 //提取图片为document ，然后用于后面判断
@@ -247,8 +290,11 @@ public class FenpanService {
                 if (o.getName().endsWith(".pdf")) {
                     Pieces pieces = new Pieces(FileTypeEnum.PDF, o.getAbsolutePath(), o.getName());
                     allDirectory.add(pieces);
-                } else if (o.getName().endsWith(".docx") || o.getName().endsWith(".doc")) {
+                } else if (o.getName().endsWith(".docx") ) {
-                    Pieces pieces = new Pieces(FileTypeEnum.WORD, o.getAbsolutePath(), o.getName());
+                    Pieces pieces = new Pieces(FileTypeEnum.DOCX, o.getAbsolutePath(), o.getName());
+                    allDirectory.add(pieces);
+                }  else if ( o.getName().endsWith(".doc")) {
+                    Pieces pieces = new Pieces(FileTypeEnum.DOC, o.getAbsolutePath(), o.getName());
                     allDirectory.add(pieces);
                 } else if (o.getName().endsWith(".jpg") || o.getName().endsWith(".png")
                         || o.getName().endsWith(".jpeg") || o.getName().endsWith(".tif")

src/main/java/com/docus/sw/fenpan/FileTypeEnum.java

@@ -1,7 +1,8 @@
 package com.docus.sw.fenpan;
 
 public enum FileTypeEnum {
-    WORD,
+    DOC,
+    DOCX,
     PDF,
     JPG
 }

src/main/java/com/docus/sw/word/GetPicsDocx.java

@@ -1,15 +1,15 @@
 package com.docus.sw.word;
 
+import org.apache.poi.xwpf.extractor.XWPFWordExtractor;
+import org.apache.poi.xwpf.usermodel.XWPFDocument;
+import org.apache.poi.xwpf.usermodel.XWPFPictureData;
+
 import java.io.File;
 import java.io.FileInputStream;
 import java.io.FileOutputStream;
 import java.io.IOException;
 import java.util.List;
 
-import org.apache.poi.xwpf.extractor.XWPFWordExtractor;
-import org.apache.poi.xwpf.usermodel.XWPFDocument;
-import org.apache.poi.xwpf.usermodel.XWPFPictureData;
-
 public class GetPicsDocx {
     public static void main(String[] args) {
         String path = "E:\\上海项目测试\\文档\\35.docx";
@@ -32,4 +32,35 @@ public class GetPicsDocx {
             e.printStackTrace();
         }
     }
+
+
+    public static void getPics(String fromPath, String toPath) {
+        File file = new File(fromPath);
+        FileInputStream fis = null;
+        try {
+            fis = new FileInputStream(file);
+            XWPFDocument document = new XWPFDocument(fis);
+            XWPFWordExtractor xwpfWordExtractor = new XWPFWordExtractor(document);
+            String text = xwpfWordExtractor.getText();
+            System.out.println(text);
+            List<XWPFPictureData> picList = document.getAllPictures();
+            int i = 1;
+            for (XWPFPictureData pic : picList) {
+                byte[] bytev = pic.getData();
+                FileOutputStream fos = new FileOutputStream(toPath +i+ pic.getFileName());
+                fos.write(bytev);
+            }
+
+        } catch (IOException e) {
+            e.printStackTrace();
+        } finally {
+            if (fis != null) {
+                try {
+                    fis.close();
+                } catch (IOException e) {
+                    e.printStackTrace();
+                }
+            }
+        }
+    }
 }

src/main/java/com/docus/sw/word/ReadImgDoc.java

@@ -1,15 +1,17 @@
 package com.docus.sw.word;
 
-import java.io.*;
-import java.util.*;
-
-
 import org.apache.poi.hwpf.HWPFDocument;
 import org.apache.poi.hwpf.model.PicturesTable;
 import org.apache.poi.hwpf.usermodel.CharacterRun;
 import org.apache.poi.hwpf.usermodel.Picture;
 import org.apache.poi.hwpf.usermodel.Range;
 
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.FileOutputStream;
+import java.io.OutputStream;
+import java.util.UUID;
+
 public class ReadImgDoc {
 
     public static void main(String[] args) throws Exception {
@@ -40,4 +42,28 @@ public class ReadImgDoc {
 
     }
 
+
+    private static void readPicture(String path, String toPath) throws Exception {
+        FileInputStream in = new FileInputStream(new File(path));
+        HWPFDocument doc = new HWPFDocument(in);
+        int length = doc.characterLength();
+        PicturesTable pTable = doc.getPicturesTable();
+        // int TitleLength=doc.getSummaryInformation().getTitle().length();
+
+        //  System.out.println(TitleLength);
+        // System.out.println(length);
+        for (int i = 0; i < length; i++) {
+            Range range = new Range(i, i + 1, doc);
+            CharacterRun cr = range.getCharacterRun(0);
+            if (pTable.hasPicture(cr)) {
+                Picture pic = pTable.extractPicture(cr, false);
+                String afileName = pic.suggestFullFileName();
+                OutputStream out = new FileOutputStream(new File(toPath + i + afileName));
+                pic.writeImageContent(out);
+            }
+        }
+
+    }
+
+
 }