Physical path issue... asp.NET 2.0 - VB

[gchq]

I want to use images located outside of the Application - and thought I was being quite smart by including two paths in Global.asax, one for the Server and one for the build machine that VS2005 sits on - then used this bit of code so that it would run on either...

Private Function ReturnPath()
        Dim vImagePath As String
        Dim vPath As New Uri(Request.Url.ToString)
        Dim vPathString As String = Left(vPath.Authority, 9)
        If vPathString = "localhost" Then
            vImagePath = gFilePath.gFilePathLocal
        Else
            vImagePath = gFilePath.gFilePathServer
 
        End If
        Return vImagePath
 
    End Function
 

Now I must be really missing something here, because I assumed that the path would always be relative to the machine it was being run on, so that on the server it would be C:\Somefolder and on my workstation it would map to K:\Somefolder... But no!! If I load the app onto the server and open it in a browser, it's still wanting the path from my local machine.. So, how do I map to a file outside of the App so that the server locates it, not the local machine? Talk about confuzzed....

[Plater]

Have you debugged to see if your other IF condition is ever met?

[gchq]

Hi Plater

Indeed the other condition is being met and that is the whole problem! I've just unearthed a revelation - I've been writing this stuff since the early days of (classic) asp and only now realised that if you put in a physical path it will map to the physical path of the client machine NOT the server...

So the question is - I guess - is there any method to map to a directory that is not only outside of the app, but also outside of InetPub?

I would assume the only way would be to map a virtual directory from IIS first and then find a method although I have a feeling that .NET security will have an issue with even that....

[Plater]

Well you could code a "serve up" page (for lack of a better term)
I use it to serve up PDF files.
I have a directory on Computer A where a lot of company documents are kept.
ComputerB has a webserver for our use. I use an intermediate page (and a little more, but don't worry about that) to acces the files on ComputerA without mapping a virtual directory.

So I would have say:
http://ComputerB/PDFServe.aspx?ProductGuidelines.pdf

And PDFServe.aspx looks to see if that file exists, if it does, it clears out the Response (Response.Clear() ) and then sets the contenttype to that of a PDF and sends the file bytes down.

Employee's here may have no idea where these documents are physically stored, nor have access to them, beyond what I provide in the webpage functionality.

You can do the same for images, or anything you want. Using backend code, you are not bound to the locale created by the web application under the inetpub directory.

[gchq]

So how does 'PDFServe.aspx' on ComputerB access files on Computer A?

I can't quite figure out the dynamics of once the page gets the Request.QueryString (for ProductGuidlines.pdf) how it aquires the file...

The other issue is.once the Directory is located it has to load all the file details into a JavaScript array

 
....  "var fadeimages=new Array(); " & _
        "while (i <=" & IndexMax & ") { " & _
        "vNext=(vImage+i); " & _
        "if((vNext >= 10)) { " & _
        "NextPic= target02 + vNext; } " & _
        "else if((vNext > 99)) { " & _
        "NextPic = target03 + vNext; } " & _
        "else if((vNext <= 9)) { " & _
        "NextPic = target01 + vNext; }" & _
        "fadeimages[+i]=['" & vPath & Folder & "/" & FileStart & "' + NextPic + '.jpg', '', '']; " & _
        "i = i+1; " & _ ......
 

[Plater]

In my page_load() event for the serve up page:

string getfile = ""; 
try
{
   getfile = @"\\ComputerA\pdfs\";
   //add in the file name you want, i.e.:
   // getfile+="SomePDF.pdf";
   if ((!(new System.IO.FileInfo(getfile).Exists)))
   {
      ReportError("The system could not find the requested file for SiteID: " + siteid);
   }
   else
   {
      Response.Clear();
      Response.Cache.SetRevalidation(HttpCacheRevalidation.AllCaches);
      Response.ContentType = "Application/PDF";
      Response.WriteFile(getfile);
      Response.End();
   }
}
catch (Exception ee)
{
   if (ee.Message.ToLower().StartsWith("thread was being aborted"))
   {//ignore, should probably catch ThreadAbortException
   }
   else
   {
      ReportError("An error has occured: '" + ee.Message + "' ");
   }
}
 

You should be able to get some idea from that.


Then when you link, instead of doing say this:

<img src="/images/myimage.jpg" alt="" />
 

You would say:

<img src="/ServeImage.aspx?Image=myimage.jpg" alt="" />
 

[gchq]

Had to scratch my head regarding the server mapping for a bit. but with:-

 
GetFile = "\\Interclaims1\InterclaimsC\backup\sk\002\01.jpg"
 
 

It worked a treat! Thank you so much!

I have another issue that this might really help with - I have web apps that produce pdf files on the fly and setting up new templates and installing them on each app looked like a nightmare - with this concept I should be able to have just one template (per item) and instead of :-

 
 rtnInput = TK.OpenInputFile(Server.MapPath("PDFInput\nucs-4072.pdf"))
 
 

I could just reference the one storage place for the pdf template - even offer differing templates based on database selection! Now I'm off.....

Thanks again