我有一个ASP.NET SOAP Web服务，它的Web方法创建一个PDF文件，将其写入应用程序的“下载”目录，并将URL返回给用户。 代码：
//Create the map images (MapPrinter) and insert them on the PDF (PagePrinter).
MemoryStream mstream = null;
FileStream fs = null;
try
{
//Create the memorystream storing the pdf created.
mstream = pgPrinter.GenerateMapImage();
//Convert the memorystream to an array of bytes.
byte[] byteArray = mstream.ToArray();
//return byteArray;
//Save PDF file to site's Download folder with a unique name.
System.Text.StringBuilder sb = new System.Text.StringBuilder(Global.PhysicalDownloadPath);
sb.Append("\\");
string fileName = Guid.NewGuid().ToString() + ".pdf";
sb.Append(fileName);
string filePath = sb.ToString();
fs = new FileStream(filePath, FileMode.CreateNew);
fs.Write(byteArray, 0, byteArray.Length);
string requestURI = this.Context.Request.Url.AbsoluteUri;
string virtPath = requestURI.Remove(requestURI.IndexOf("Service.asmx")) + "Download/" + fileName;
return virtPath;
}
catch (Exception ex)
{
throw new Exception("An error has occurred creating the map pdf.", ex);
}
finally
{
if (mstream != null) mstream.Close();
if (fs != null) fs.Close();
//Clean up resources
if (pgPrinter != null) pgPrinter.Dispose();
}
然后在Web服务的Global.asax文件中，我在Application_Start事件监听器中设置了一个定时器。在定时器的ElapsedEvent监听器中，我查找下载目录中任何早于定时器间隔（测试为1分钟，部署约20分钟）的文件并将其删除。 代码：
//Interval to check for old files (milliseconds), also set to delete files older than now minus this interval.
private static double deleteTimeInterval;
private static System.Timers.Timer timer;
//Physical path to Download folder. Everything in this folder will be checked for deletion.
public static string PhysicalDownloadPath;
void Application_Start(object sender, EventArgs e)
{
// Code that runs on application startup
deleteTimeInterval = Convert.ToDouble(System.Configuration.ConfigurationManager.AppSettings["FileDeleteInterval"]);
//Create timer with interval (milliseconds) whose elapse event will trigger the delete of old files
//in the Download directory.
timer = new System.Timers.Timer(deleteTimeInterval);
timer.Enabled = true;
timer.AutoReset = true;
timer.Elapsed += new System.Timers.ElapsedEventHandler(OnTimedEvent);
PhysicalDownloadPath = System.Web.Hosting.HostingEnvironment.ApplicationPhysicalPath + "Download";
}
private static void OnTimedEvent(object source, System.Timers.ElapsedEventArgs e)
{
//Delete the files older than the time interval in the Download folder.
var folder = new System.IO.DirectoryInfo(PhysicalDownloadPath);
System.IO.FileInfo[] files = folder.GetFiles();
foreach (var file in files)
{
if (file.CreationTime < DateTime.Now.AddMilliseconds(-deleteTimeInterval))
{
string path = PhysicalDownloadPath + "\\" + file.Name;
System.IO.File.Delete(path);
}
}
}
这个功能完美地运行，只有一个例外。当我将 Web 服务应用程序发布到 inetpub\wwwroot（Windows 7，IIS7）时，它不会删除 Download 目录中的旧文件。当我从不在 wwwroot 中的物理目录发布到 IIS 时，该应用程序运行得很完美。显然，IIS 对 web 根目录中的文件施加了某种锁定。我已经测试过模拟管理员用户运行应用程序，但仍然无法正常工作。有没有什么方法可以在编程时规避 wwwroot 中的锁定？客户可能希望将应用程序发布到根目录。