67 lines
3.3 KiB
HTML
67 lines
3.3 KiB
HTML
|
<html>
|
||
|
|
<head>
|
||
|
|
<link rel="stylesheet" type="text/css" href="..\styles.css">
|
||
|
|
</head>
|
||
|
|
<body>
|
||
|
|
<img src="..\..\ro.png" alt="" align="right">
|
||
|
|
|
||
|
|
<!-------------------------------------------------------------------------->
|
||
|
|
|
||
|
|
<p class="h1"><span>Extended File Transfer Sample</span></p>
|
||
|
|
|
||
|
|
<h2>Purpose</h2>
|
||
|
|
|
||
|
|
<p>
|
||
|
|
Some time ago we received several questions in our newsgroup asking how to download
|
||
|
|
big files using the RemObjects SDK. This sample presents a working solution.
|
||
|
|
</p>
|
||
|
|
|
||
|
|
<h2>How it works</h2>
|
||
|
|
<p>
|
||
|
|
The approach is quite simple,instead of downloading a complete file at once, it
|
||
|
|
is split into small chunks to be downloaded separately. On the client side, these are then combined back into a single file.
|
||
|
|
</p>
|
||
|
|
<p>
|
||
|
|
This sample provides a RemObject server that exposes the content of a folder (see the <b>GetFilesList</b> method in the ExtendedFileTransferService implementation)
|
||
|
|
and a client that can connect to the server to obtain a list of the files available for downloading. The server also provides another method for downloading a chunk
|
||
|
|
of a file called <b>DownloadFilePart.</b>
|
||
|
|
This method requires three parameters:
|
||
|
|
</p>
|
||
|
|
<ul>
|
||
|
|
<li><b>FileName</b> (AnsiString) - name of the file to be downloaded.</li>
|
||
|
|
<li><b>PartNo</b> (int32) - the chunk part that should be downloaded starting with 1. </li>
|
||
|
|
<li><b>PartSize</b> (int32) - size of of chunk that should be downloaded. </li>
|
||
|
|
</ul>
|
||
|
|
<strong>DownloadFilePart </strong>reads the section of the file with offset <b>PartSize*PartNo</b> and returns the data as a Binary stream to the client.
|
||
|
|
|
||
|
|
When we want to download a file we can specify size of chunk (Below the <b>Chunk Size</b> combobox you will see calculated total count of chunks).
|
||
|
|
Also we can specify the number of
|
||
|
|
threads we want to use for the downloading operation. Since each partial chunk download
|
||
|
|
is independent, using threads is quite attractive in this scenario.
|
||
|
|
|
||
|
|
|
||
|
|
<h2>Examine the code</h2>
|
||
|
|
<ul>
|
||
|
|
<li>See the implementation of the <b>GetFilesList</b> and <b>DownloadFilePart</b> methods in the <b>ExtendedFileTransferService_Impl</b> class. Look how we can read a chunk of file and return it back to the client.</li>
|
||
|
|
<li>See <b>DownloadPart</b> function on the client side. This is the method that executes inside each thread we run to download the selected file. Note that we use a separate channel and message for each thread.</li>
|
||
|
|
<li>Review the <b>bDownload_Click</b> handler. Here we are starting the download. Look how we are using a thread pool for controlling the number of the threads that are
|
||
|
|
involved in the downloading process.</li>
|
||
|
|
</ul>
|
||
|
|
|
||
|
|
|
||
|
|
|
||
|
|
<h2>Getting started</h2>
|
||
|
|
<ul>
|
||
|
|
<li>Compile and run the server and client applications.</li>
|
||
|
|
<li>Put a file into shared folder (you can easily locate and open this folder when you
|
||
|
|
click on the link label on the server main form).</li>
|
||
|
|
<li>Get the available files on the client side by clicking the <b>GetFiles</b> button.</li>
|
||
|
|
<li>Adjust the chunk size and thread count on the client.</li>
|
||
|
|
<li>Try to download the selected file. At the end of downloading process, you can open
|
||
|
|
this file to ensure that downloading was successful. </li>
|
||
|
|
</ul>
|
||
|
|
|
||
|
|
<!-------------------------------------------------------------------------->
|
||
|
|
</body>
|
||
|
|
</html>
|