如何使用 blueimp jquery 上传插件获取每个块的 md5 迭代 md5 总和

How to get md5 iterated md5 sum of every chunk, using blueimp jquery upload plugin

我需要计算并发送一个迭代的 md5-hash 到我的上传-api。 但是我不知道怎么办。

我正在使用此处找到的教程:

http://tutorialzine.com/2013/05/mini-ajax-file-upload-form/

连同 blueimp jquery 上传插件。

仅发送一个文件(文件大小小于块大小),一切正常。但是,如果一个文件被分块,那么我不知道如何捕获该块以获取它的 md5。

最后,我必须像这里描述的那样迭代地制作 md5:

https://code.google.com/p/crypto-js/#Progressive_Hashing

$('#upload').fileupload({

    // This element will accept file drag/drop uploading
    dropZone: $('#drop'),

    type        : GLOBAL_FORM_METHOD,
    method      : "post",   // Type of data-send-method
    dataType    : "json",   // Type of data to recieve from api-call
    maxChunkSize: GLOBAL_CHUNK_SIZE,
    multipart   : true,

    // This function is called when a file is added to the queue;
    // either via the browse button, or via drag/drop:
    add: function (e, data)
    {
        var reader = new FileReader();
        var file = data.files[0];
        var jqXHR;

        var tpl = $('<li class="working"><input type="text" value="0" data-width="48" data-height="48"'+
            ' data-fgColor="#0788a5" data-readOnly="1" data-bgColor="#3e4043" /><p></p><div class="msg"></div><span></span></li>');

        // Append the file name and file size
        tpl.find('p')
            .text( file.name )
            .append('<i>' + formatFileSize( file.size ) + '</i>');

        // Add the HTML to the UL element
        data.context = tpl.appendTo(ul);

        // Initialize the knob plugin
        tpl.find('input').knob();

        // Listen for clicks on the cancel icon
        tpl.find('span').click(function()
        {
            if( tpl.hasClass('working') )
            {
                jqXHR.abort();
            }

            tpl.fadeOut( function()
            {
                tpl.remove();
            });
        });

        // Prevent XHR from sending data in "multipart/formData"
        data.postMessage = data.files[0].type;
        data.contentType = data.files[0].type;

        var chunksize = GLOBAL_CHUNK_SIZE > file.size ? file.size : GLOBAL_CHUNK_SIZE;

        // Describe the FileReader-DataLoad-Event
        reader.onload = function( event ) 
        {
            var binary = event.target.result;
            var md5 = CryptoJS.MD5(binary).toString();

            data.url += "&md5sum=" + md5;

            // D A T A   S E N D 
            jqXHR = data.submit();
        };

        // ADD url to XHR-object
        data.url = GLOBAL_FORM_ACTION;
        data.url += "?etf_id=" + GLOBAL_FOLDER_ID;
        data.url += "&file_title=" + file.name;

        // If the file will be send in one piece...
        if( GLOBAL_CHUNK_SIZE > file.size )
        {
            // ADD url-parameter to XHR-object
            data.url += "&size_chunk_start=" + 0;
            data.url += "&size_chunk_length=" + chunksize;
        }
        // This part for the chunks must be in "beforeSend"-Callback,
        // because, the chunk-related size-data is undefined in this case
        // but available there.

        // ADD url-parameter to XHR-object
        data.url += "&size_final=" + file.size;

        // Read md5-sum and send the file / chunk...
        // On multipart "file" is a chunk !
        reader.readAsBinaryString( file );
    },

    beforeSend : function(e, data)
    {
        var file = data.files[0];

        this.find(".msg").hide();

        // If the file will be send as chunks...
        if( GLOBAL_CHUNK_SIZE < file.size )
        {
            console.log( "Chunk data: ", data.uploadedBytes, data.chunkSize, file.size, data );

            // ADD url-parameter to XHR-object
            data.url += "&size_chunk_start=" + data.uploadedBytes;
            data.url += "&size_chunk_length=" + data.chunkSize;

            if( typeof this.attr('session_id') !== "undefined" )
                data.url += "&session_id=" + this.attr( 'session_id' );
        }

    });

希望大家帮忙,我自己解决。

This answer to a similar question 提供了一种实现此目的的方法。还有一些库可以为你做这件事,比如 CryptoJS

要点是用 0 字节填充文件以达到可被块长度整除的特定长度,然后将数据逐块读入缓冲区,并获得该数据。然后,为您读取的每个块附加该散列。