java实现文件切片上传百度云+断点续传的方法

前言：

本文代码通过dubbo进行远程调用的接口，如果不使用dubbo，直接将service放到你的service，并稍作修改，redis替换成自己封装的工具即可。下方代码有点多，但基本上都加注释了。

切片上传百度云和上传到本地有所不同，上传到本地是需要前端给个唯一标识，通过自定义唯一标识保存到本地，最后在合并文件。而上传百度云则需要直接把切片传到百度云，唯一标识uploadId是百度云提供的，上传及合并都是直接调用百度云提供的接口。下方为百度云的使用文档。

对象存储BOS - 文件管理 | 百度智能云文档 (baidu.com)

1.定义一个实体类

用于接收前台传过来的信息，信息应该包括总切片数，当前切片数，文件名称，文件总大小。

import com.geely.tcm.common.annotation.Excel;
import lombok.Data;

import javax.validation.constraints.Max;
import javax.validation.constraints.Min;
import javax.validation.constraints.NotNull;
import java.io.Serializable;

@Data
public class MutilFileInfo implements Serializable {
    private static final long serialVersionUID = 1L;

    @NotNull(message = "车型不能为空")
    @Excel(name = "车型")
    private String model;

    @Excel(name = "文件名称")
    private String name;

    @Excel(name = "文件总大小")
    private Long totalSize;

    @Min(value = 1, message = "不能小于1")
    @Max(value = 10000, message = "不能大于10000")
    @Excel(name = "当前分片序号")
    private Integer chunk;

    @Min(value = 1, message = "不能小于1")
    @Max(value = 10000, message = "不能大于10000")
    @Excel(name = "分片总数")
    private Integer chunks;
}

2.创建controller

用于处理前端过来的文件及信息

import com.alibaba.fastjson.JSON;
import com.baidubce.services.bos.model.PartETag;
import com.geely.tcm.common.annotation.Log;
import com.geely.tcm.common.bean.MutilFileInfo;
import com.geely.tcm.common.bean.baidu.CustomPartETag;
import com.geely.tcm.common.bean.measure.Cmm3dModel;
import com.geely.tcm.common.configuration.RedisUtils;
import com.geely.tcm.common.enums.BusinessType;
import com.geely.tcm.common.util.StringUtils;
import com.geely.tcom.remote.api.baidu.BaiduCloudBosService;
import io.swagger.annotations.Api;
import io.swagger.annotations.ApiImplicitParam;
import io.swagger.annotations.ApiImplicitParams;
import io.swagger.annotations.ApiOperation;
import lombok.extern.slf4j.Slf4j;
import org.apache.dubbo.config.annotation.DubboReference;
import org.springframework.web.bind.annotation.PostMapping;
import org.springframework.web.bind.annotation.RequestMapping;
import org.springframework.web.bind.annotation.RequestParam;
import org.springframework.web.bind.annotation.RestController;
import org.springframework.web.multipart.MultipartFile;

import java.util.*;
import java.util.concurrent.Semaphore;
import java.util.stream.Collectors;

@Api(tags = "文件上传下载服务")
@RequestMapping("/threeModel")
@RestController
@Slf4j
public class Model3DController1  {
    //DubboReference这个是远程调用，如果没有远程调用，直接Autowired就可以
    @DubboReference(version = "${sheet.service.version}")
    private BaiduCloudBosService baiduCloudBosService;
    //通过加锁保证线程安全
    private Semaphore lock = new Semaphore(1);

    //api开头的这些是swagger用到的，如果没使用swagger可以删到PostMapping上边
    @ApiOperation(value = "obj模型上传")
    @ApiImplicitParams({
            @ApiImplicitParam(name = "totalSize", value = "文件总大小",dataTypeClass = Long.class),
            @ApiImplicitParam(name = "chunks", value = "总切片数",dataTypeClass = Integer.class),
            @ApiImplicitParam(name = "chunk", value = "当前切片数",dataTypeClass = Integer.class),
            @ApiImplicitParam(name = "model", value = "车型",dataTypeClass = String.class),
            @ApiImplicitParam(name = "file", value = "切片文件",dataTypeClass = MultipartFile.class)
    })
    @Log(title = "obj模型上传", businessType = BusinessType.INSERT)
    @PostMapping("/upload3DToBos")
    public Object upload3DToBos( MutilFileInfo fileinfo, @RequestParam(required=false,value="file") MultipartFile multipartFile) {
        String model=fileinfo.getModel().trim();
        Map<String,Object> m=new HashMap();
        String fileName=fileinfo.getName();
        if (StringUtils.isEmpty(fileName)) {//判断文件名称是否为空
            m.put("key","error");
            m.put("value","文件名不能为空");
            return m;
        }
        if (multipartFile == null) {//判断上传过来的切片是否为空
            m.put("key","error");
            m.put("value","请上传文件");
            return m;
        }
        try{
            String filePath="/test/"+model+"/"+fileName;//拼接百度云地址
            String redisKey="partETags-"+model+"-"+fileName;//缓存的键
            Integer current =fileinfo.getChunk();//当前切片数
            Integer total = fileinfo.getChunks();//总切片数
            Long size=multipartFile.getSize();//文件大小
            String uploadId = getUploadId(redisKey, filePath);//该方法获取百度云的唯一标识uploadId
            byte[] file = multipartFile.getBytes();//将切片过来的文件转成字节数组
            if(current==1){//这块进行断点续传，如果是第一个切片，查询一下缓存中是否有数据了，如果有返回给前台，如果没有进行上传
                if(getCurrentEqualsOne(redisKey).size()>0) {
                    //给前台提供是否缓存中已有数据
                    return getCurrentEqualsOne(redisKey);
                }
            }
            //切片上传百度云
            List<CustomPartETag> customPartETags = getPartEtags(redisKey, uploadId, filePath, file, size, current);
            //如果缓存中成功的切片数等于总数，那么进行合并
            if (customPartETags.size() == total) {
                customPartETags = customPartETags.stream().sorted(Comparator.comparing(PartETag::getPartNumber)).collect(Collectors.toList());//进行排序
                baiduCloudBosService.completePart(uploadId, customPartETags, filePath);//执行合并
                RedisUtils.hdel(redisKey,"uploadId","customPartETags");//如果合并成功，删除缓存中的数据
                // 入库做记录
                m.put("key","success");
                m.put("value","上传成功");
                return m;
            }
            m.put("key","continue");
            m.put("value",current+1);
            return m;
        }catch (Exception e){
            e.printStackTrace();
            m.put("key","error");
            m.put("value","上传失败");
            return m;
        }
    }

    /**
     * 获取 part
     *
     * @param redisPartKey redisKey
     * @return
     */
    private List<CustomPartETag> getPartEtags(String redisPartKey, String uploadId, String filePath,  byte[] file, long size, int current) {
        List<CustomPartETag> customPartETags = null;
        try {
            lock.acquire();
            //因为每次上传成功后都会把成功的切片存进百度云，如注解下方第四行代码，所以每次上传之前需要先取出来，将成功的切片放进去
            customPartETags = RedisUtils.hget(redisPartKey, "customPartETags") == null ? new ArrayList<>() : JSON.parseArray((String) RedisUtils.hget(redisPartKey, "customPartETags"), CustomPartETag.class);
            //上传百度云成功之后会返回一个PartETagg对象，由于项目中存在dubbo，顾将CustomPartETag类继承PartETagg进行序列化
            CustomPartETag customPartETag = baiduCloudBosService.uploadCustomPartETag(uploadId, filePath, file, size, current);
            customPartETags.add(customPartETag);
            setPartEtagsToRedis(redisPartKey, customPartETags);

        } catch (InterruptedException e) {
            e.printStackTrace();
        } finally {
            lock.release();
        }
        return customPartETags;
    }

    /**
     * 获取 uploadId
     *
     * @param redisPartKey redisKey
     * @param filePath     filePath
     * @return
     */
    private String getUploadId(String redisPartKey, String filePath) {
        String uploadId = null;
        try {
            lock.acquire();//线程锁
            uploadId = (String) RedisUtils.hget(redisPartKey, "uploadId");//先去redis中找，是否已经有uploadId
            if (StringUtils.isEmpty(uploadId)) {//判断是否为空，如果为空，说明之前没上传过，为首次上传，如果已经存在，返回该uploadId
                uploadId = baiduCloudBosService.initMultipartUpload(filePath);//进行初始化，返回一个唯一标识uploadId
                RedisUtils.hset(redisPartKey, "uploadId", uploadId, 60 * 60);//将返回来的uploadId放到缓存中，下个切片上传直接使用这个uploadId
            }
        } catch (InterruptedException e) {
            e.printStackTrace();
        } finally {
            lock.release();
        }
        return uploadId;
    }

    /**
     * 薄墩 partKey到redis
     *
     * @param redisPartKey
     * @param customPartETags
     */
    private void setPartEtagsToRedis(String redisPartKey, List<CustomPartETag> customPartETags) {
        RedisUtils.hset(redisPartKey, "customPartETags", JSON.toJSONString(customPartETags), 60 * 60);
    }
    public Map<String,Object> getCurrentEqualsOne(String redisKey){
        Map<String,Object> m=new HashMap<>();
        List<CustomPartETag>  customPartETags = RedisUtils.hget(redisKey, "customPartETags") == null ? new ArrayList<>() : JSON.parseArray((String) RedisUtils.hget(redisKey, "customPartETags"), CustomPartETag.class);
        //给前台提供是否缓存中已有数据
        if(customPartETags.size()==0){
            return m;
        }else{
                //缓存中有数据,返回缓存中数据
                int t=  customPartETags.get(customPartETags.size()-1).getPartNumber();
                m.put("key","hasValue");
                m.put("value",t);
                return  m;
            }
    }
}

3.远程的service

(如果不用远程调用，直接创建一个service就可以)

package com.geely.tcm.sheet.service;

import com.baidubce.services.bos.BosClient;
import com.baidubce.services.bos.model.*;
import com.geely.tcm.common.bean.baidu.BosResult;
import com.geely.tcm.common.bean.baidu.CustomPartETag;
import com.geely.tcm.common.util.StringUtils;
import com.geely.tcom.remote.api.baidu.BaiduCloudBosService;
import lombok.extern.slf4j.Slf4j;
import org.apache.dubbo.config.annotation.DubboService;
import org.springframework.beans.factory.annotation.Autowired;

import java.io.*;
import java.net.URL;
import java.util.ArrayList;
import java.util.List;

/**
 * 百度私有云服务操作
 *
 * @author : wangbin
 * @date : 2021/9/24
 * @since : 1.0.0
 */
@Slf4j
@DubboService(version = "${sheet.service.version}")
public class BaiduCloudBosServiceImpl{

    @Autowired
    private BosClient bosClient;

    public String initMultipartUpload(String filePath) {
        // 开始Multipart Upload
        InitiateMultipartUploadRequest initiateMultipartUploadRequest = new InitiateMultipartUploadRequest(null, filePath);
        InitiateMultipartUploadResponse initiateMultipartUploadResponse = bosClient.initiateMultipartUpload(initiateMultipartUploadRequest);
        // 打印UploadId，它是区分分块上传事件的唯一标识
        String uploadId = initiateMultipartUploadResponse.getUploadId();
        log.info("filePath= {},uploadId = {}", filePath, uploadId);
        return uploadId;
    }

    public CustomPartETag uploadCustomPartETag(String uploadId, String filePath, byte[] file, long partSize, int partNumber) {
        UploadPartRequest uploadPartRequest = new UploadPartRequest();
        uploadPartRequest.setBucketName(null);
        uploadPartRequest.setKey(filePath);
        uploadPartRequest.setUploadId(uploadId);
        InputStream input= new ByteArrayInputStream(file);
        uploadPartRequest.setInputStream(input);
        uploadPartRequest.setPartSize(partSize);
        uploadPartRequest.setPartNumber(partNumber );
        uploadPartRequest.setMd5Digest("");
        UploadPartResponse uploadPartResponse = bosClient.uploadPart(uploadPartRequest);
        PartETag partETag = uploadPartResponse.getPartETag();
        CustomPartETag  customPartETag = new CustomPartETag();
        customPartETag.setETag(partETag.getETag());
        customPartETag.setPartNumber(partETag.getPartNumber());
        return customPartETag;
    }

    /**
     * 分块上传结束之后，当所有的数据Part验证通过后，BOS将把这些数据part组合成一个完整的Object
     *
     * @param uploadId
     * @param customPartETags
     * @param filePath
     */
    public void completePart(String uploadId, List<CustomPartETag> customPartETags, String filePath) {
        //for循环转一下 CustomPartETag 转PartETag
        if(customPartETags.size()>0){
            List<PartETag> partETags=new ArrayList<>();
            for(CustomPartETag c:customPartETags){
                PartETag partETag=new PartETag();
                partETag.setETag(c.getETag());
                partETag.setPartNumber(c.getPartNumber());
                partETags.add(partETag);
            }
            CompleteMultipartUploadRequest completeMultipartUploadRequest = new CompleteMultipartUploadRequest(null, filePath, uploadId, partETags);
            CompleteMultipartUploadResponse completeMultipartUploadResponse = bosClient.completeMultipartUpload(completeMultipartUploadRequest);
            log.info("完成分片上传 ， filePath = {},uploadId = {}, etag = {}", filePath, uploadId,completeMultipartUploadResponse.getETag());
        }
    }
}

4.定义一个实体类

继承，实现序列化，（dubbo不序列化就会传不过去）

import com.baidubce.services.bos.model.PartETag;
import java.io.Serializable;

public class CustomPartETag extends PartETag implements Serializable {
}

5.这里提供一个临时测试的前端代码

<html lang="en">
<head>
    <meta charset="UTF-8">
    <title>upload</title>
    <script src="https://cdn.bootcss.com/jquery/3.2.1/jquery.min.js"></script>
</head>
<body>
    <input type="file" name="file" id="file">
    <button id="upload" onClick="upload()">upload</button>
    <script type="text/javascript">
        var bytesPerPiece = 1024 * 1024; // 每个文件切片大小定为1MB .
        var totalPieces;

        function upload() {
            var blob = document.getElementById("file").files[0];
            var start = 0;
            var end;
            var index = 1;
            var filesize = blob.size;
            var filename = blob.name;

            //计算文件切片总数
            totalPieces = Math.ceil(filesize / bytesPerPiece);
            while(start < filesize) {
                end = start + bytesPerPiece;
                if(end > filesize) {
                    end = filesize;
                }

                var chunk = blob.slice(start,end);//切割文件
                var sliceIndex= blob.name + index;
                var formData = new FormData();
                formData.append("file", chunk, filename);
    formData.append("totalSize",filesize);
    formData.append("chunks",totalPieces);
    formData.append("chunk",index);
    formData.append("model",'kx11');
                $.ajax({
//路径换成你的路径
                    url: 'http://127.0.0.1:6300/threeModel/upload3DToBos',
                    type: 'POST',
     dataType:'json',
                    cache: false,
                    data: formData,
                    processData: false,
                    contentType: false,
     mimeType: "multipart/form-data",
     success: function (result){
      console.log(result.value);
     }
    });

                start = end;
                index++;
            }
        }
    </script>
</body>
</html>

到此这篇关于java实现文件切片上传百度云+断点续传的方法的文章就介绍到这了,更多相关java 文件切片上传 内容请搜索我们以前的文章或继续浏览下面的相关文章希望大家以后多多支持我们！