Re: Url rewriting using IHttpHandlerFactory

IMO you don't need to use RewritePath. RewritePath is full of problems man.
First being that it totally breaks postbacks.

I'm not sure if it's too late in the request or not as I've never used an
HttpHandler to rewrite paths. I've always used an HttpModule to use
RewritePath.

Doing the rewrite in your HttpHandlerFactory is mimicking what ASP.NET does
but, you have more control. And postbacks actually work.

Hope that helps. I'm really tired and not that coherent right now. ;)

-----Original Message-----
From: Discussion of building .NET applications targeted for the Web
[mailto:DOTNET-WEB@xxxxxxxxxxxxxxxxxxx] On Behalf Of Andreas H åkansson
Sent: Thursday, June 09, 2005 1:53 AM
To: DOTNET-WEB@xxxxxxxxxxxxxxxxxxx
Subject: Re: [DOTNET-WEB] Url rewriting using IHttpHandlerFactory

Chris,

 So.. would this by any chance mean that it's not possible, ie. perhaps to
late in  the request life-cycle, to rewrite the path from inside a custom
HttpHandler returned by a HttpHandlerFactory?

This wouldn't actually be a problem - I'd just write my own interface for
interpreting variosu requests, so I can extend with additional formats
later on - however I would like my above "theory" confirmed or corrected
so I know what is going on.

Then again, your solution doesn't use the rewritepath at all.. how is it
to be used in such a scenario then?

Thanks,

.andreas

On Wed, 8 Jun 2005 16:46:01 -0700, Chris Martin <ghettoblaster@xxxxxxx>
wrote:

>        If (TargetPage.Length = 0) Then
>            Return PageParser.GetCompiledPageInstance(url, pathTranslated,
>context)
>        Else
>                Return PageParser.GetCompiledPageInstance(url,
>Server.MapPath(GetCorrectPage(url)),
>context)
>        End If
>
>-----Original Message-----
>From: Discussion of building .NET applications targeted for the Web
>[mailto:DOTNET-WEB@xxxxxxxxxxxxxxxxxxx] On Behalf Of Andreas H åkansson
>Sent: Wednesday, June 08, 2005 11:29 AM
>To: DOTNET-WEB@xxxxxxxxxxxxxxxxxxx
>Subject: [DOTNET-WEB] Url rewriting using IHttpHandlerFactory
>
>Hi,
>
>I'm playing around with url rewriting and I have come across a problem
>which
>I
>can't seem to get past. The general ide is to have a IHttpHandlerFactory
>class
>which checks the incoming request url against a set of predefined rules
and
>creates and return the appropriate IHttpHandler implementation.
>
>Inside the IHttpHandler implementations, the request is examined and
>interpreted
>in the correct way. In short they parse a request into a new url (for
>example, could
>also be a handler to generate an rss-feed or whatever, hence the need for
>the
>factory and set of rules) and rewrites it using context.RewritePath()
>
>My problem is (verified by stepping the debugger) is that the factory
>creates the
>correct Handler.. and the andler interprets the request and calls
>RewritePath()
>but the page is not displayed in my browser, instead a blank page is
shown.
>
>Basicly this is the setup of the testsite (used to reproduce the problem)
>is
>as
>follows
>
>    Root
>        Test
>            one.aspx
>            two.aspx
>        Webform.aspx
>
>The URL Rewriting should be able to take Root/Test/1.aspx and rewrite it
>into
>Root/Test/one.aspx and the same for 2/two.aspx - so it's a very basic test
>setup.
>
>The handler which is created it setup as following
>
>// CustomHandler.vb
>Public Class CustomHandler
>    Implements IHttpHandler
>
>    Public ReadOnly Property IsReusable() As Boolean Implements
>System.Web.IHttpHandler.IsReusable
>        Get
>            Return False
>        End Get
>    End Property
>
>    Public Sub ProcessRequest(ByVal context As System.Web.HttpContext)
>Implements System.Web.IHttpHandler.ProcessRequest
>
>        Dim TargetPage As String = _
>            Regex.Match(context.Request.Url.AbsolutePath, "[0-9]+\.aspx",
>RegexOptions.IgnoreCase).Value
>
>        If (Not TargetPage Is Nothing) Then
>            Select Case TargetPage
>                Case "1.aspx"
>                    context.RewritePath("~/test/one.aspx")
>                Case "2.aspx"
>                    context.RewritePath("~/test/two.aspx")
>            End Select
>        End If
>
>    End Sub
>
>End Class
>
>The factory implementation (remember it's not bulletproof, it's a test app
>to reproduce the issue at hand)
>
>Public Class HttpRewriter
>    Implements IHttpHandlerFactory
>
>    Public Function GetHandler(ByVal context As System.Web.HttpContext,
>ByVal requestType As String, ByVal url As String, ByVal pathTranslated As
>String) As System.Web.IHttpHandler Implements
>System.Web.IHttpHandlerFactory.GetHandler
>
>        Dim TargetPage As String = _
>            Regex.Match(context.Request.Url.AbsolutePath, "[0-9]+\.aspx",
>RegexOptions.IgnoreCase).Value
>
>        If (TargetPage.Length = 0) Then
>            Return PageParser.GetCompiledPageInstance(url, pathTranslated,
>context)
>        End If
>
>        Return New CustomHandler
>
>    End Function
>
>    Public Sub ReleaseHandler(ByVal handler As System.Web.IHttpHandler)
>Implements System.Web.IHttpHandlerFactory.ReleaseHandler
>
>    End Sub
>
>End Class
>
>the factory has been added to web.config and all is fine.. When debugging
>I
>can see that when I enter
>http://localhost/myapp/test/1.aspx the factory reaches the "Return New
>CustomHandler" line and execution
>is shifted over to the new objects ProcessRequest method. Continuing to
>step
>the code and I see it hits the
>Case "1.aspx" in the select block, as expected and calls
>context.RewritePath("~/test/one.aspx")
>
>This is when I'm left with a blank page. I have my sucpicions, i.e there
>is
>no compiled instance of the real
>page (one.aspx) returned from my handler (and I can't since it's a sub) ..
>how would .NET know how to
>render my page then? I've seen this design before in the .Text blog engine
>source code, but I've been
>unable to get it to work.
>
>Please remember that this is a basic test setup used to reproduce the
>problem, in hope that someone can
>point how the (obvious?) misstake I'm making. I need for a factory to
>create
>different handlers based on
>the requested path, so I can't do it like in a few articles I've seen
>where
>they rewrite the path and call
>GetCompiledPageInstance .. both done in the factory ..
>
>I figure my handler is correct, but the problem is in my factory.. perhaps
>I
>need to do something to my
>handler object before I return it in the factory?
>
>Any advice, suggestions, pointers, help etc is apprechiated! Don't worry
>about the VB.NET part i code
>C# as well so don't feel hindred by the code :P
>
>Thanks!
>
>===================================
>This list is hosted by DevelopMentor®  http://www.develop.com
>
>View archives and manage your subscription(s) at
http://discuss.develop.com
>
>===================================
>This list is hosted by DevelopMentor®  http://www.develop.com
>
>View archives and manage your subscription(s) at
http://discuss.develop.com

===================================
This list is hosted by DevelopMentor®  http://www.develop.com

View archives and manage your subscription(s) at http://discuss.develop.com

===================================
This list is hosted by DevelopMentor®  http://www.develop.com

View archives and manage your subscription(s) at http://discuss.develop.com