如何使用webclient异步下载多个文件，但一次只能一个？

Question

It has been surprisingly hard to find a code example of downloading multiple files using the webclient class asynchronous method, but downloading one at a time.

How can I initiate a async download, but wait until the first is finished until the second, etc. Basically a que.

(note I do not want to use the sync method, because of the increased functionality of the async method.)

The below code starts all my downloads at once. (the progress bar is all over the place)

private void downloadFile(string url)
        {
            WebClient client = new WebClient();

            client.DownloadProgressChanged += new DownloadProgressChangedEventHandler(client_DownloadProgressChanged);
            client.DownloadFileCompleted += new AsyncCompletedEventHandler(client_DownloadFileCompleted);

            // Starts the download
            btnGetDownload.Text = "Downloading...";
            btnGetDownload.Enabled = false;
            progressBar1.Visible = true;
            lblFileName.Text = url;
            lblFileName.Visible = true;
            string FileName = url.Substring(url.LastIndexOf("/") + 1,
                            (url.Length - url.LastIndexOf("/") - 1));
             client.DownloadFileAsync(new Uri(url), "C:\\Test4\\" + FileName);

        }

        void client_DownloadFileCompleted(object sender, AsyncCompletedEventArgs e)
        {

        }

        void client_DownloadProgressChanged(object sender, DownloadProgressChangedEventArgs e)
        {
            double bytesIn = double.Parse(e.BytesReceived.ToString());
            double totalBytes = double.Parse(e.TotalBytesToReceive.ToString());
            double percentage = bytesIn / totalBytes * 100;
            progressBar1.Value = int.Parse(Math.Truncate(percentage).ToString());
        }

Answer 1

What I have done is populate a Queue containing all my urls, then I download each item in the queue. When there are no items left, I can then process all the items. I've mocked some code up below. Keep in mind, the code below is for downloading strings and not files. It shouldn't be too difficult to modify the below code.

    private Queue<string> _items = new Queue<string>();
    private List<string> _results = new List<string>();

    private void PopulateItemsQueue()
    {
        _items.Enqueue("some_url_here");
        _items.Enqueue("perhaps_another_here");
        _items.Enqueue("and_a_third_item_as_well");

        DownloadItem();
    }

    private void DownloadItem()
    {
        if (_items.Any())
        {
            var nextItem = _items.Dequeue();

            var webClient = new WebClient();
            webClient.DownloadStringCompleted += OnGetDownloadedStringCompleted;
            webClient.DownloadStringAsync(new Uri(nextItem));
            return;
        }

        ProcessResults(_results);
    }

    private void OnGetDownloadedStringCompleted(object sender, DownloadStringCompletedEventArgs e)
    {
        if (e.Error == null && !string.IsNullOrEmpty(e.Result))
        {
            // do something with e.Result string.
            _results.Add(e.Result);
        }
        DownloadItem();
    }

Edit: I've modified your code to use a Queue. Not entirely sure how you wanted progress to work. I'm sure if you wanted the progress to cater for all downloads, then you could store the item count in the 'PopulateItemsQueue()' method and use that field in the progress changed method.

    private Queue<string> _downloadUrls = new Queue<string>();

    private void downloadFile(IEnumerable<string> urls)
    {
        foreach (var url in urls)
        {
            _downloadUrls.Enqueue(url);
        }

        // Starts the download
        btnGetDownload.Text = "Downloading...";
        btnGetDownload.Enabled = false;
        progressBar1.Visible = true;
        lblFileName.Visible = true;

        DownloadFile();
    }

    private void DownloadFile()
    {
        if (_downloadUrls.Any())
        {
            WebClient client = new WebClient();
            client.DownloadProgressChanged += client_DownloadProgressChanged;
            client.DownloadFileCompleted += client_DownloadFileCompleted;

            var url = _downloadUrls.Dequeue();
            string FileName = url.Substring(url.LastIndexOf("/") + 1,
                        (url.Length - url.LastIndexOf("/") - 1));

            client.DownloadFileAsync(new Uri(url), "C:\\Test4\\" + FileName);
            lblFileName.Text = url;
            return;
        }

        // End of the download
        btnGetDownload.Text = "Download Complete";
    }

    private void client_DownloadFileCompleted(object sender, AsyncCompletedEventArgs e)
    {
        if (e.Error != null)
        {
            // handle error scenario
            throw e.Error;
        }
        if (e.Cancelled)
        {
            // handle cancelled scenario
        }
        DownloadFile();
    }

    void client_DownloadProgressChanged(object sender, DownloadProgressChangedEventArgs e)
    {
        double bytesIn = double.Parse(e.BytesReceived.ToString());
        double totalBytes = double.Parse(e.TotalBytesToReceive.ToString());
        double percentage = bytesIn / totalBytes * 100;
        progressBar1.Value = int.Parse(Math.Truncate(percentage).ToString());
    }

Answer 2

I am struggling to understand where the issue is. If you only call the async method for the first file, won't it only download that file? Why not use the client_downlaodFileCompleted event to initiate the next file download based on some value passed by AsyncCompletedEvents, or maintain a list of downloaded files as a static variable and have client_DownloadFileCompleted iterate tht list to find the next file to download.

Hope that helps but please post more information if I have miunderstood your question.

Answer 3

我会创建一个新方法，例如名为getUrlFromQueue，它从队列（集合或数组）返回一个新url并删除它..然后它调用downloadFile（url） - 并在client_DownloadFileCompleted中再次调用getUrlFromQueue。