我正在使用jquery filedrop，我在Internet Explorer（10+）上使用readAsBinaryString收到错误。它适用于Chrome和Firefox。
然后我将该方法更改为readAsDataURL，这在每个浏览器中都有效。
问题是最后一个方法从HttpPostedFileBase返回一个base64文件，我不知道如何保存文件。
现在是我的代码：
byte[] binaryData;
binaryData = new Byte[file.InputStream.Length];
long bytesRead = file.InputStream.Read(binaryData, 0, (int)file.InputStream.Length);
file.InputStream.Close();
string base64String = System.Convert.ToBase64String(binaryData, 0, binaryData.Length);
byte[] bytes = System.Convert.FromBase64String(base64String);
string filePath = Path.Combine(ed_fisico_provisorio.vr_parametro, no_arquivo_provisorio);
System.IO.FileStream fs = System.IO.File.Create(filePath);
fs.Write(bytes, 0, bytes.Length);
fs.Close();
但是像这样保存，文件的内容就像
数据：应用程序/ msword; base64,0M8R4KGxGuEAAAAAAAAAAAAAAAAAAAAAPgADAP7 ...
还有其他想法吗？
-----更新-----
逐步澄清正在发生的事情。顺便说一下，我正在使用C＃MVC。
第1步 - 使用插件jquery.filedrop.js，我将文件放在其视图中。可以是任何文件扩展名。
第2步 - 调用插件，其中包含以下代码：
// Respond to an upload
function upload() {
stop_loop = false;
if (!files) {
opts.error(errors[0]);
return false;
}
var filesDone = 0,
filesRejected = 0;
if (files_count > opts.maxfiles && opts.queuefiles === 0) {
opts.error(errors[1]);
return false;
}
// Define queues to manage upload process
var workQueue = [];
var processingQueue = [];
var doneQueue = [];
// Add everything to the workQueue
for (var i = 0; i < files_count; i++) {
workQueue.push(i);
}
// Helper function to enable pause of processing to wait
// for in process queue to complete
var pause = function (timeout) {
setTimeout(process, timeout);
return;
}
// Process an upload, recursive
var process = function () {
var fileIndex;
if (stop_loop) return false;
// Check to see if are in queue mode
if (opts.queuefiles > 0 && processingQueue.length >= opts.queuefiles) {
return pause(opts.queuewait);
} else {
// Take first thing off work queue
fileIndex = workQueue[0];
workQueue.splice(0, 1);
// Add to processing queue
processingQueue.push(fileIndex);
}
try {
if (beforeEach(files[fileIndex]) != false) {
if (fileIndex === files_count) return;
var reader = new FileReader(),
max_file_size = 1048576 * opts.maxfilesize;
reader.index = fileIndex;
if (files[fileIndex].size > max_file_size) {
opts.error(errors[2], files[fileIndex], fileIndex);
// Remove from queue
processingQueue.forEach(function (value, key) {
if (value === fileIndex) processingQueue.splice(key, 1);
});
filesRejected++;
return true;
}
reader.onloadend = send;
//reader.readAsArrayBuffer(files[fileIndex]);
reader.readAsDataURL(files[fileIndex]);
//reader.readAsBinaryString(files[fileIndex]);
} else {
filesRejected++;
}
} catch (err) {
// Remove from queue
processingQueue.forEach(function (value, key) {
if (value === fileIndex) processingQueue.splice(key, 1);
});
opts.error(errors[0]);
return false;
}
// If we still have work to do,
if (workQueue.length > 0) {
process();
}
};
var send = function (e) {
var fileIndex = ((typeof (e.srcElement) === "undefined") ? e.target : e.srcElement).index
// Sometimes the index is not attached to the
// event object. Find it by size. Hack for sure.
if (e.target.index == undefined) {
e.target.index = getIndexBySize(e.total);
}
var xhr = new XMLHttpRequest(),
upload = xhr.upload,
file = files[e.target.index],
index = e.target.index,
start_time = new Date().getTime(),
boundary = '------multipartformboundary' + (new Date).getTime(),
builder;
newName = rename(file.name);
mime = file.type
if (typeof newName === "string") {
builder = getBuilder(newName, e.target.result, mime, boundary);
} else {
builder = getBuilder(file.name, e.target.result, mime, boundary);
}
upload.index = index;
upload.file = file;
upload.downloadStartTime = start_time;
upload.currentStart = start_time;
upload.currentProgress = 0;
upload.startData = 0;
upload.addEventListener("progress", progress, false);
xhr.open("POST", opts.url, true);
xhr.setRequestHeader('content-type', 'multipart/form-data; boundary=' + boundary);
// Add headers
$.each(opts.headers, function (k, v) {
xhr.setRequestHeader(k, v);
});
xhr.sendAsBinary(builder);
opts.uploadStarted(index, file, files_count);
xhr.onload = function () {
if (xhr.responseText) {
var now = new Date().getTime(),
timeDiff = now - start_time,
result = opts.uploadFinished(index, file, jQuery.parseJSON(xhr.responseText), timeDiff, xhr);
filesDone++;
// Remove from processing queue
processingQueue.forEach(function (value, key) {
if (value === fileIndex) processingQueue.splice(key, 1);
});
// Add to donequeue
doneQueue.push(fileIndex);
if (filesDone == files_count - filesRejected) {
afterAll();
}
if (result === false) stop_loop = true;
}
};
}
// Initiate the processing loop
process();
}
第3步 - 插件返回到我收到文件的控制器并尝试保存：
[HttpPost]
public ActionResult UploadFiles(IEnumerable<HttpPostedFileBase> files)
{
byte[] binaryData;
binaryData = new Byte[file.InputStream.Length];
long bytesRead = file.InputStream.Read(binaryData, 0, (int)file.InputStream.Length);
file.InputStream.Close();
string base64String = System.Convert.ToBase64String(binaryData, 0, binaryData.Length);
byte[] bytes = System.Convert.FromBase64String(base64String);
//Salva arquivo físico em pasta provisória
string filePath = Path.Combine(ed_fisico_provisorio.vr_parametro, file.FileName);
System.IO.FileStream fs = System.IO.File.Create(filePath);
fs.Write(bytes, 0, bytes.Length);
fs.Close();
}
保存文件的内容如上所述。
任何帮助将不胜感激。
答案 0 :(得分：1)
有点不清楚您使用什么技术来接收文件（WEBAPI，MVC，Something Awesome），但我过去在WebAPI 2端点中使用过它，并使用流处理文件数据。
public async Task<IHttpActionResult> Post()
{
var provider = new MultipartMemoryStreamProvider();
var stream = await Request.Content.ReadAsMultipartAsync(provider).ConfigureAwait(false);
// Returns the first file it finds
var fileStream = stream.Contents.FirstOrDefault(x => x.Headers.ContentDisposition.FileName != null);
// return a result
}
使用fileStream对象，如果文件内容是文本，则可以使用fileStream.ReadAsStreamAsync()或fileStream.ReadAsStringAsync()。