文件上传根据编码读取txt文件内容重新写入UTF-8编码Txt文件

àì夳堔傛蜴生んèń 2023-06-25 03:16 5阅读 0赞

nginx转发地址为 fileUploadUrl

package com.cncbox.saoye.test;
    
    import lombok.extern.log4j.Log4j2;
    import org.apache.commons.lang3.StringUtils;
    import org.springframework.beans.factory.annotation.Value;
    import org.springframework.stereotype.Component;
    import org.springframework.web.multipart.MultipartFile;
    
    import javax.annotation.PostConstruct;
    import java.io.*;
    import java.util.ArrayList;
    import java.util.List;
    
    /**
     * @author : zhang sq
     * @date : 2019/12/25 9:38
     **/
    @Component
    @Log4j2
    public class FileUtils {
    
        //nginx转发请配置此地址    /=C:/ftp/root/
        @Value("${file.update.url}")
        private String fileUploadUrl; //yml 配置文件上传地址  例: C:/ftp/root/
    
        @Value("${server.url}")
        private String serverName;  //服务名    saoye
    
        //文件上传根路径
        public static String filePath;
        //文件上传服务名
        private static String serverPath;
    
        @PostConstruct  //启动时候直接赋值
        public void init() {
            filePath = this.fileUploadUrl;
            serverPath = this.serverName;
            log.error("===============filePath参数赋值{}=============", filePath);
            log.error("===============serverPath参数赋值{}=============", serverPath);
        }
    
        /**
         * 批量上传
         * @param namePath 上传地址  格式例:  /新西游记/大闹天宫/斗法二郎神/
         */
        public static List<String> uploadFile(List<MultipartFile> files, String namePath) {
            if (null == files || files.size() == 0) {
                log.error("===============files集合文件为空=============");
                return null;
            }
            List<String> fileList = new ArrayList<>();
            for (MultipartFile file : files) {
                String uploadFile = uploadFile(file, namePath);
                if (null == uploadFile) {
                    log.error("===============files集合文件有内容为空或错误=============");
                    return null;
                }
                fileList.add(uploadFile);
            }
            return fileList;
        }
    
        /**
         * 文件上传
         * @param namePath   格式例:  /新西游记/大闹天宫/斗法二郎神/
         * @param file 文件
         * @return 上传地址   返回为null  证明参数有问题
         */
        public static String uploadFile(MultipartFile file, String namePath) {
            log.info("=====file值为:{}=====namePath值为:{}",file,namePath);
            //文件为空
            if (null == file || StringUtils.isBlank(file.getOriginalFilename())) {
                log.error("===============文件为空=============");
                return null;
            }
            String fileName = file.getOriginalFilename();//文件名  abc.png
            //文件名为空 或者格式错误
            if (!fileName.contains(".") || StringUtils.isEmpty(fileName)) {
                log.error("===============文件格式错误,文件全名为{}=============", fileName);
                return null;
            }
            //将字符全部小写截取文件后缀  txt   png  img
            String fileType = fileName.toLowerCase().substring(fileName.lastIndexOf(".") + 1);
            try {
                StringBuffer ssb = new StringBuffer(); //上传地址
                long timeMillis = System.currentTimeMillis();
                int random = (int) (Math.random() * (100) + 100);
                StringBuffer sb = new StringBuffer();  //返回地址
                //创建文件夹  saoye/史记/史记四十七/txt/12345632.txt
                sb=sb.append(serverPath.toLowerCase())
                        .append(namePath).append(fileType).append("/").append(String.valueOf(timeMillis))
                        .append(String.valueOf(random)).append(".").append(fileType);
                //  c:/ftp/root/saoye/史记/史记四十七/txt/12345632.txt
                String fileDir = ssb.append(filePath.toLowerCase()).append(sb).toString();
    
                File dirFile = new File(fileDir);
                if (!dirFile.exists()) {
                    dirFile.mkdirs();
                }
                file.transferTo(dirFile);
                log.info("===========上传真实地址为: {}",dirFile);
                log.info("============上传返回地址为 {}=====",sb);
                return sb.toString();
            } catch (Exception e) {
                e.printStackTrace();
            }
            return null;
        }
    
        /**
         * 删除图片 1->删除成功  0->删除失败
         * @param fileUrl 文件路径
         */
        public static int deleteFileByUrl(String fileUrl) {
            log.info("============传入删除文件地址为 {}=====",fileUrl);
            StringBuffer sb=new StringBuffer();
            fileUrl=sb.append(filePath.toLowerCase()).append(fileUrl).toString();
            log.info("============文件真实删除地址为 {}=====",fileUrl);
            File file = new File(fileUrl);
            int result = 0;
            if (file.exists()) {
                file.delete();
                log.info("=================={}:文件已删除", fileUrl);
                return 1;
            }
            return result;
        }
    
        /**
         * 读取txt文件内容
         * @param fileEncod  文件编码格式 UTF-8/GBK
         * @param txtPath 文件路径
         */
        public static String readTxtFile(String txtPath,String fileEncod) {
            log.info("=======传入文件路径为:{}  , 文件编码格式为:{}",txtPath,fileEncod);
            StringBuffer sb=new StringBuffer();
            txtPath=sb.append(filePath.toLowerCase()).append(txtPath).toString();
            log.info("================读取文件真实路径为{}", txtPath);
            if (StringUtils.isEmpty(txtPath)) {
                log.error("================传入路径错误为null:{}", txtPath);
                return null;
            }
            String fileType = txtPath.toLowerCase().substring(txtPath.lastIndexOf(".") + 1);
            if (!fileType.equals("txt")) {
                log.error("================传入文件格式错误,不是txt文件 :{}", txtPath);
                return null;
            }
            File dirFile = new File(txtPath);
            if (!dirFile.exists()) {
                log.error("================传入文件路径错误,文件不存在 :{}", txtPath);
                return null;
            }
            String fileContent = "";
            try {
                InputStreamReader read = new InputStreamReader(new FileInputStream(txtPath), fileEncod);
                BufferedReader reader = new BufferedReader(read);
                String line;
                while ((line = reader.readLine()) != null) {
                    fileContent += line;
                }
                read.close();
            } catch (Exception e) {
                e.printStackTrace();
                log.error("=========文件读取错误==========");
                throw new RuntimeException();
            }
            return fileContent;
        }
    
        /**
         * @param filePostfix    写入文件地址  格式      /西游记/三打白骨精/第四章/
         * @param fileContent 写入文件内容
         * @return 写入文件地址
         */
        public static String writeTxtFile(String filePostfix, String fileContent) {
            StringBuffer sb=new StringBuffer(); //返回地址
            int random = (int) (Math.random() * (100) + 100); //113
            sb=sb.append(serverPath.toLowerCase()).append(filePostfix); //
            StringBuffer fileName=new StringBuffer(); //文件名称
            fileName=fileName.append(System.currentTimeMillis()).append(String.valueOf(random)).append(".txt");
            StringBuffer ssb=new StringBuffer(); //写入地址
            String fileUrl=ssb.append(filePath.toLowerCase()).append(sb).toString();
            String filUrlpath=filePath.toLowerCase()+sb+fileName; //文件最终存储真实地址
            try {
                File dir = new File(fileUrl);
                if (!dir.exists()) {
                    dir.mkdirs();
                }
                File file = new File(filUrlpath);
                file.createNewFile();
                log.info("===============新的写入地址为:{}",filUrlpath);
                OutputStreamWriter write = new OutputStreamWriter(new FileOutputStream(file), "utf-8");
                BufferedWriter writer = new BufferedWriter(write);
                writer.write(fileContent);
                writer.close();
            } catch (Exception e) {
                e.printStackTrace();
                log.error("===============文件写入错误:文件地址:{}   文件内容: {}",filUrlpath, fileContent);
                deleteFileByUrl(sb.append(fileName).toString()); //删除错误文件
                throw new RuntimeException(e);
            }
           return sb.append(fileName).toString();
        }
    
        //判断txt文件编码格式方法
        public static  String getFilecharset(File sourceFile) {
            String charset = "GBK";
            byte[] first3Bytes = new byte[3];
            try {
                boolean checked = false;
                BufferedInputStream bis = new BufferedInputStream(new FileInputStream(sourceFile));
                bis.mark(0);
                int read = bis.read(first3Bytes, 0, 3);
                if (read == -1) {
                    return charset; //文件编码为 ANSI
                } else if (first3Bytes[0] == (byte) 0xFF
                        && first3Bytes[1] == (byte) 0xFE) {
                    charset = "UTF-16LE"; //文件编码为 Unicode
                    checked = true;
                } else if (first3Bytes[0] == (byte) 0xFE
                        && first3Bytes[1] == (byte) 0xFF) {
                    charset = "UTF-16BE"; //文件编码为 Unicode big endian
                    checked = true;
                } else if (first3Bytes[0] == (byte) 0xEF
                        && first3Bytes[1] == (byte) 0xBB
                        && first3Bytes[2] == (byte) 0xBF) {
                    charset = "UTF-8"; //文件编码为 UTF-8
                    checked = true;
                }
                bis.reset();
                if (!checked) {
                    int loc = 0;
                    while ((read = bis.read()) != -1) {
                        loc++;
                        if (read >= 0xF0)
                            break;
                        if (0x80 <= read && read <= 0xBF) // 单独出现BF以下的，也算是GBK
                            break;
                        if (0xC0 <= read && read <= 0xDF) {
                            read = bis.read();
                            if (0x80 <= read && read <= 0xBF) // 双字节 (0xC0 - 0xDF)
                                // (0x80
                                // - 0xBF),也可能在GB编码内
                                continue;
                            else
                                break;
                        } else if (0xE0 <= read && read <= 0xEF) {// 也有可能出错，但是几率较小
                            read = bis.read();
                            if (0x80 <= read && read <= 0xBF) {
                                read = bis.read();
                                if (0x80 <= read && read <= 0xBF) {
                                    charset = "UTF-8";
                                    break;
                                } else
                                    break;
                            } else
                                break;
                        }
                    }
                }
                bis.close();
            } catch (Exception e) {
                e.printStackTrace();
            }
            log.info("===============文件编码格式为: {}",charset);
            return charset;
        }
    
        public static void main(String[] args) {
            String filepath="C:\\ftp\\root\\saoye\\李白\\诗仙\\第六页\\1576639675444135.txt";
            String file = writeTxtFile("/西游记/三打白骨精/第四章/", "文件编码格式为");
            System.out.println(file);
        }
    }