doc网上专用接口

This commit is contained in:
Sakura9701
2022-06-08 10:34:36 +08:00
parent 0b75952a35
commit b32f5dbdaa
7 changed files with 379 additions and 148 deletions

View File

@@ -0,0 +1,71 @@
package com.example.ts_obj.Util;
import javax.servlet.http.HttpServletResponse;
import java.io.IOException;
import java.io.InputStream;
import java.io.OutputStream;
import java.net.HttpURLConnection;
import java.net.MalformedURLException;
import java.net.URL;
public class FileDownloadUtil {
/**
* @return
* @description: 从服务器获得一个输入流
*/
public static InputStream getInputStream(String urlPath) {
InputStream inputStream = null;
HttpURLConnection httpURLConnection = null;
try {
URL url = new URL(urlPath);
httpURLConnection = (HttpURLConnection) url.openConnection();
// 设置网络连接超时时间
httpURLConnection.setConnectTimeout(3000);
// 设置应用程序要从网络连接读取数据
httpURLConnection.setDoInput(true);
httpURLConnection.setRequestMethod("GET");
int responseCode = httpURLConnection.getResponseCode();
System.out.println("responseCode is:" + responseCode);
if (responseCode == HttpURLConnection.HTTP_OK) {
// 从服务器返回一个输入流
inputStream = httpURLConnection.getInputStream();
} else {
inputStream = httpURLConnection.getErrorStream();
}
} catch (MalformedURLException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
return inputStream;
}
/**
* @param resp
* @param inputStream
* @description: 将输入流输出到页面
*/
public static void writeFile(HttpServletResponse resp, InputStream inputStream) {
OutputStream out = null;
try {
out = resp.getOutputStream();
int len = 0;
byte[] b = new byte[1024];
while ((len = inputStream.read(b)) != -1) {
out.write(b, 0, len);
}
out.flush();
} catch (IOException e) {
e.printStackTrace();
} finally {
try {
if (out != null) {
out.close();
}
} catch (Exception e) {
e.printStackTrace();
}
}
}
}

View File

@@ -0,0 +1,76 @@
package com.example.ts_obj.Util;
import com.github.jaiimageio.plugins.tiff.TIFFImageWriteParam;
import javax.imageio.*;
import javax.imageio.metadata.IIOMetadata;
import javax.imageio.stream.FileImageInputStream;
import javax.imageio.stream.ImageInputStream;
import javax.imageio.stream.ImageOutputStream;
import java.awt.image.BufferedImage;
import java.io.File;
import java.io.IOException;
import java.util.Iterator;
import java.util.Locale;
/**
* 识别图片中的文字
*
* @author zlj
*
*/
public class ImageIOHelper {
/**
* 创建临时图片文件
*
* @param imageFile
* @return
* @throws IOException
*/
public File createImage(File imageFile) throws IOException {
Iterator<ImageReader> readers = ImageIO.getImageReaders(new FileImageInputStream(imageFile));
ImageReader reader = readers.next();
ImageInputStream iis = ImageIO.createImageInputStream(imageFile);
reader.setInput(iis);
IIOMetadata streamMetadata = reader.getStreamMetadata();
TIFFImageWriteParam tiffWriteParam = new TIFFImageWriteParam(Locale.CHINESE);
tiffWriteParam.setCompressionMode(ImageWriteParam.MODE_DISABLED);
Iterator<ImageWriter> writers = ImageIO.getImageWritersByFormatName("tiff");
ImageWriter writer = writers.next();
BufferedImage bi = reader.read(0);
IIOImage image = new IIOImage(bi, null, reader.getImageMetadata(0));
File tempFile = tempImageFile(imageFile);
ImageOutputStream ios = ImageIO.createImageOutputStream(tempFile);
writer.setOutput(ios);
writer.write(streamMetadata, image, tiffWriteParam);
ios.close();
iis.close();
writer.dispose();
reader.dispose();
return tempFile;
}
/**
* 添加后缀 tempfile
*
* @param imageFile
* @return
* @throws IOException
*/
private File tempImageFile(File imageFile) throws IOException {
String path = imageFile.getPath();
StringBuffer strB = new StringBuffer(path);
strB.insert(path.lastIndexOf('.'), "_text_recognize_temp");
String s = strB.toString().replaceFirst("(?<=//.)(//w+)$", "tif");
Runtime.getRuntime().exec("attrib " + "\"" + s + "\"" + " +H"); // 设置文件隐藏
return new File(strB.toString());
}
}

View File

@@ -1,14 +1,20 @@
package com.example.ts_obj.controller;
import com.example.ts_obj.Util.*;
import com.spire.doc.*;
import com.spire.doc.Document;
import com.spire.doc.collections.CellCollection;
import com.spire.doc.collections.RowCollection;
import com.spire.doc.collections.SectionCollection;
import com.spire.doc.collections.TableCollection;
import com.spire.doc.documents.*;
import com.spire.doc.fields.DocPicture;
import com.alibaba.fastjson.JSON;
import com.deepoove.poi.XWPFTemplate;
import com.deepoove.poi.config.Configure;
import com.deepoove.poi.data.*;
import com.deepoove.poi.data.style.Style;
import com.deepoove.poi.policy.ParagraphRenderPolicy;
import com.example.ts_obj.Util.CustomTableRenderPolicy;
import com.example.ts_obj.Util.Util;
import com.example.ts_obj.Util.XWPFUtils;
import com.example.ts_obj.bean.ReturnCodeAndMsgEnum;
import com.example.ts_obj.bean.ReturnValue;
import com.example.ts_obj.entity.Typeset;
@@ -19,12 +25,14 @@ import com.example.ts_obj.service.TypesetService;
import com.example.ts_obj.service.UserService;
import io.swagger.annotations.Api;
import io.swagger.annotations.ApiOperation;
import jdk.nashorn.internal.runtime.regexp.joni.Regex;
import org.apache.commons.collections4.CollectionUtils;
import org.apache.commons.lang3.StringUtils;
import org.apache.poi.openxml4j.util.ZipSecureFile;
import org.apache.poi.util.Units;
import org.apache.poi.xwpf.usermodel.*;
import org.apache.xmlbeans.XmlCursor;
import org.openxmlformats.schemas.drawingml.x2006.wordprocessingDrawing.CTInline;
import org.openxmlformats.schemas.wordprocessingml.x2006.main.*;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
@@ -33,9 +41,15 @@ import org.springframework.util.FileCopyUtils;
import org.springframework.web.bind.annotation.*;
import org.springframework.web.multipart.MultipartFile;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpServletResponse;
import java.awt.*;
import java.io.*;
import java.text.SimpleDateFormat;
import java.util.*;
import java.util.List;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
@RestController
@RequestMapping("/typeset")
@@ -44,8 +58,10 @@ public class TypesetController {
private final static Logger logger = LoggerFactory.getLogger(UserController.class);
private String BASE_DIR="E:/upload/";
private static String NEW_BASE_DIR="E:/upload/20220520/";
private String BASE_DIR="C:\\Users\\Administrator\\Desktop\\typesetcode\\upload\\";
private static String NEW_BASE_DIR="C:\\Users\\Administrator\\Desktop\\typesetcode\\upload\\20220520\\";
private static String IMG_DIR="C:\\Users\\Administrator\\Desktop\\typesetcode\\pictures\\";
// private String BASE_DIR = "/home/wwwroot/ts.tmrjournals.com/upload/";
@Autowired
private TypesetService typesetservice;
@@ -386,6 +402,49 @@ public class TypesetController {
return new ReturnValue(ReturnCodeAndMsgEnum.Success, al);
}
//网上专用文件读取接口
@ApiOperation(value = "网上专用文件读取接口")
@PostMapping("webReaddoc")
public ReturnValue DocReader(String fileRoute, HttpServletRequest request, HttpServletResponse response) {
ArrayList al = new ArrayList();
try {
InputStream inputStream = FileDownloadUtil.getInputStream(fileRoute);
// FileDownloadUtil.writeFile(response, inputStream);
XWPFDocument xdoc = new XWPFDocument(inputStream);
List<XWPFParagraph> list = xdoc.getParagraphs();
String cache = "";
String cach = "";
for (XWPFParagraph para:list) {
List<XWPFRun> runs = para.getRuns();
for (XWPFRun r : runs) {
System.out.println("run====>"+r);
if (r.getText(0) == null) {
continue;
}
if (r.isItalic() && r.isBold()) {
cach = "<b><i>" + r.getText(0) + "</i></b>";
} else if (r.isItalic()) {
cach = "<i>" + r.getText(0) + "</i>";
} else if (r.isBold()) {
cach = "<b>" + r.getText(0) + "</b>";
} else {
cach = r.getText(0);
}
cache += cach;
}
}
al.add(cache);
} catch (Exception e) {
e.printStackTrace();
}
return new ReturnValue(ReturnCodeAndMsgEnum.Success, al);
}
@PostMapping("/mytest")
public ReturnValue mytest(Long typesetId) {
@@ -728,7 +787,7 @@ public class TypesetController {
//遍历照片
// for (XWPFPictureData picture: allPictures) {
byte[] data = pictureData.getData();
File filePicture = new File("C:\\Users\\yl\\Desktop\\pictures\\" + imageName.replace("image",""));
File filePicture = new File(IMG_DIR + imageName.replace("image",""));
if (!filePicture.exists()){
if (filePicture.getParentFile().exists()) {
filePicture.getParentFile().mkdirs();
@@ -751,31 +810,157 @@ public class TypesetController {
XWPFDocument newXdoc = new XWPFDocument(newFis);
List<XWPFParagraph> paragraphs = newXdoc.getParagraphs();
for (int i=0 ; i < paragraphs.size() ; i++){
System.out.println("内容为"+paragraphs.get(i).getRuns());
List<XWPFRun> runs = paragraphs.get(i).getRuns();
String replace = String.valueOf(runs).replace(",", "").replace(" ","");
for (String pic: picNameList){
if (replace.contains(picMap.get(pic).replace(",","").replace(" ",""))){
System.out.println("包含执行了");
//获取光标
XmlCursor xmlCursor = paragraphs.get(i+1).getCTP().newCursor();
XWPFParagraph xwpfParagraph = newXdoc.insertNewParagraph(xmlCursor);
XWPFRun run = xwpfParagraph.createRun();
run.addBreak(BreakClear.ALL);
// run.setText("{{@image}}");
FileInputStream inputStream = new FileInputStream("C:\\Users\\yl\\Desktop\\pictures\\"+pic.replace("image",""));
run.addPicture(inputStream,XWPFDocument.PICTURE_TYPE_JPEG,pic, Units.toEMU(500),Units.toEMU(180));
//
//其他方式
Document doc = new Document();
doc.loadFromFile(this.NEW_BASE_DIR + newFileName);
Section sec = doc.getSections().get(0);
System.out.println("sec=================>"+sec);
String text = doc.getText();
System.out.println("text=================>"+text);
Paragraph para = sec.getParagraphs().get(0);
System.out.println("para=================>"+para);
for(int i = 0; i < doc.getSections().getCount(); i++) {
Section section = doc.getSections().get(i);
for (int j = 0; j < section.getParagraphs().getCount(); j++) {
Paragraph paragraph = section.getParagraphs().get(j);
System.out.println(j+"新方式内容"+paragraph.getText());
for (String pic: picNameList) {
String replace = String.valueOf(paragraph.getText()).replace(",", "").replace(" ", "");
if (replace.contains(picMap.get(pic).replace(",", "").replace(" ", ""))) {
System.out.println("包含");
String imgPath = IMG_DIR + pic.replace("image", "");
//判断图片中是否包含 文字
// String valCode = new OCRUtil().recognizeText(new File(imgPath));
// System.out.println("valCode==============>"+valCode);
// //正则表达式
// Pattern p = Pattern.compile("[\u4e00-\u9fa5]");
// Matcher m = p.matcher(valCode);
// System.out.println(m.find());
// if (m.find()){
// System.out.println("查找到了文字");
// //1.包含中文 单栏
// DocPicture docPicture = paragraph.appendPicture(imgPath);
// docPicture.setWidth(500f);
// docPicture.setHeight(180f);
// docPicture.setTextWrappingStyle(TextWrappingStyle.Top_And_Bottom);
// docPicture.setTextWrappingType(TextWrappingType.Both);
// }else {
//2.不包含 判断图片大小决定 图片为单栏还是双栏
if (pic.replace("image","").equals("1.png")){
//图片为小图 双栏
DocPicture docPicture = paragraph.appendPicture(imgPath);
docPicture.setWidth(200f);
docPicture.setHeight(180f);
docPicture.setTextWrappingStyle(TextWrappingStyle.Top_And_Bottom);
docPicture.setTextWrappingType(TextWrappingType.Both);
}else {
//图片为大图 单栏
DocPicture docPicture = paragraph.appendPicture(imgPath);
docPicture.setWidth(500f);
docPicture.setHeight(180f);
docPicture.setTextWrappingStyle(TextWrappingStyle.Top_And_Bottom);
docPicture.setTextWrappingType(TextWrappingType.Both);
}
// }
}
}
}
}
}
FileOutputStream os = new FileOutputStream("C:\\Users\\yl\\Desktop\\pictures\\New.doc");
newXdoc.write(os);
SectionCollection secs = doc.getSections();
//处理表格
for (int i = 0; i < secs.getCount(); i++) {
//获取第一个section
// Section section = doc.getSections().get(0);
TableCollection tables = secs.get(i).getTables();
System.out.println(tables);
//获取section中第一个表格
// Table table = sec.getTables().get(0);
int count = tables.getCount();
System.out.println("===================>"+count);
for (int j = 0; j < tables.getCount(); j++) {
//给表格应用样式
tables.get(j).applyStyle(DefaultTableStyle.Table_Classic_1);
// //设置表格的右边框
// tables.get(j).getTableFormat().getBorders().getRight().setBorderType(BorderStyle.Hairline);
// tables.get(j).getTableFormat().getBorders().getRight().setLineWidth(0F);
// tables.get(j).getTableFormat().getBorders().getRight().setColor(Color.white);
//
//设置表格的顶部边框
tables.get(j).getTableFormat().getBorders().getTop().setBorderType(BorderStyle.Hairline);
tables.get(j).getTableFormat().getBorders().getTop().setLineWidth(2.25F);
tables.get(j).getTableFormat().getBorders().getTop().setColor(Color.black);
//
// //设置表格的左边框
// tables.get(j).getTableFormat().getBorders().getLeft().setBorderType(BorderStyle.Hairline);
// tables.get(j).getTableFormat().getBorders().getLeft().setLineWidth(0F);
// tables.get(j).getTableFormat().getBorders().getLeft().setColor(Color.white);
//
//设置表格的底部边框
tables.get(j).getTableFormat().getBorders().getBottom().setBorderType(BorderStyle.Hairline);
// tables.get(j).getTableFormat().getBorders().getLeft().setColor(Color.black);
tables.get(j).getTableFormat().getBorders().getBottom().setLineWidth(2.25F);
//设置表格的水平和垂直边框
// tables.get(j).getTableFormat().getBorders().getVertical().setBorderType(BorderStyle.Hairline);
// tables.get(j).getTableFormat().getBorders().getHorizontal().setBorderType(BorderStyle.Hairline);
// tables.get(j).getTableFormat().getBorders().getVertical().setLineWidth(0F);
// tables.get(j).getTableFormat().getBorders().getHorizontal().setLineWidth(0F);
// tables.get(j).getTableFormat().getBorders().getVertical().setColor(Color.white);
// tables.get(j).getRows().get(0).setHeight(99F);
// tables.get(j).getRows().get(0).getRowFormat().getBorders().setBorderType(BorderStyle.Hairline);
// RowCollection rows = tables.get(j).getRows();
// for (int l = 0; l < rows.getCount(); l++) {
// rows.get(l).setHeight(15F);
// }
CellCollection rowCollection = tables.get(j).getRows().get(0).getCells();
for (int k = 0; k < rowCollection.getCount(); k++) {
rowCollection.get(k).getCellFormat().getBorders().getBottom().setBorderType(BorderStyle.Hairline);
rowCollection.get(k).getCellFormat().getBorders().getBottom().setLineWidth(15F);
// rowCollection.get(k).getCellFormat().getBorders().getBottom().setColor(Color.red);
}
// tables.get(j).getRows().get(0).getCells().get(0).getCellFormat().getBorders().getBottom().setBorderType(BorderStyle.Hairline);
tables.get(j).autoFit(AutoFitBehaviorType.Auto_Fit_To_Window);
// tables.get(j).getTableFormat().setHorizontalAlignment(RowAlignment.Right);
}
}
// for (int i=0 ; i < paragraphs.size() ; i++){
// System.out.println("内容为"+paragraphs.get(i).getRuns());
// List<XWPFRun> runs = paragraphs.get(i).getRuns();
// String replace = String.valueOf(runs).replace(",", "").replace(" ","");
// for (String pic: picNameList){
// if (replace.contains(picMap.get(pic).replace(",","").replace(" ",""))){
// System.out.println("包含执行了");
// //获取光标
// XmlCursor xmlCursor = paragraphs.get(i+1).getCTP().newCursor();
// XWPFParagraph xwpfParagraph = newXdoc.insertNewParagraph(xmlCursor);
//
// XWPFRun run = xwpfParagraph.createRun();
//// run.addBreak(BreakClear.ALL);
//// run.setText("{{@image}}");
// FileInputStream inputStream = new FileInputStream("C:\\Users\\yl\\Desktop\\pictures\\"+pic.replace("image",""));
// run.addPicture2(inputStream,XWPFDocument.PICTURE_TYPE_JPEG,pic, Units.toEMU(500),Units.toEMU(180));
// //
// }
// }
//
// }
// FileOutputStream os = new FileOutputStream(IMG_DIR+"New.doc");
// newXdoc.write(os);
doc.saveToFile(IMG_DIR+"New.doc");
System.out.println("输出完毕");
os.close();
// os.close();
fis.close();
newFis.close();
}catch (Exception e){