I'm trying to create a route that will caption a url like
http://mysite.com/tech/
but I also have an actual directory under the site that is /tech, which contains other static resources. Not my design choice, but I'm migrating an older site to mvc and dont want to break a bunch of very old links.
My route seems to keep failing, and what I can figure is that the server sees this directory and is trying to serve up the default file from it... but none is found, and it fails.
I'd like it to use my route instead and let me serve up my view. Any best practices for this?
UPDATE:
I tried using RouteExistingFiles = true, which made this work, but then I had issues with files which existed inside these directories. (for instance: http://mysite.com/tech/image.jpg). I started adding IgnoreRoute entries for all the various file types that I wanted ignored (ie .jpg files..), but this turned into a big mess. There's gotta be a better way?
You need to set the RouteExistingFiles property of your RouteCollections object to true. This will make your routing override the physical location if a clash is found.
routes.MapRoute("Default", "{controller}/{action}/{id}",
new { controller = "Home", action = "Index", id = "" }
);
routes.RouteExistingFiles = true;
To selectively ignore files/extensions you can use :
routes.IgnoreRoute("{*staticfile}", new { staticfile = #".*\.(css|js|gif|jpg)(/.*)?" });
Another approach which might help if you don't like the ignoreRoute approach, is to extend the ViewEngine and override the file exists implementaion with your own rules.
public class MyViewEngine : RazorViewEngine
{
protected override bool FileExists(ControllerContext controllerContext, string virtualPath)
{
//Some Logic to check for file
}
}
ViewEngines.Engines.Clear();
ViewEngines.Engines.Add(new CustomViewEngine());
Related
I have a Solution structure like this:
MyApp.Core
--Properties
--References
--bin
--Events
|EventHandlers.cs
--Directory
--Controllers
|DirectoryController.cs
--Helpers
|ContextHelpers.cs
--Models
|DirectoryModel.cs
--AnotherSite
--Controllers
--Helpers
--Models
--Services
--Shared
--Controllers
|HomePageController.cs
--Helpers
|Extensions.cs
|app.config
|packages.config
MyApp.Umbraco
--Properties
--References
--bin
etc........
--Views
--Directory
--Partials
|DirectoryFilters.cshtml
|DirectoryBase.cshtml
|DirectoryHome.cshtml
|FDirectory.cshtml
|SDirectory.cshtml
--Partials
--Shared
|Base.cshtml
|Web.config
etc........
My Umbraco instance uses the models and controllers from my "Core" project. There is nested directory structure, because of multiple websites in one installation, in the "Core", and also in the "Views" directory in the Umbraco instance.
I am still fairly noob to .NET MVC, and I understand route hijacking, but the documentation for Umbraco's routing is slim. I have the following:
EventHandlers.cs
namespace MyApp.Core.Events
{
/// <summary>
/// Registers site specific Umbraco application event handlers
/// </summary>
public class MyAppStartupHandler : IApplicationEventHandler
{
public void OnApplicationInitialized(UmbracoApplicationBase umbracoApplication, ApplicationContext applicationContext)
{
}
public void OnApplicationStarted(UmbracoApplicationBase umbracoApplication, ApplicationContext applicationContext)
{
RegisterCustomRoutes();
}
public void OnApplicationStarting(UmbracoApplicationBase umbracoApplication, ApplicationContext applicationContext)
{
}
private static void RegisterCustomRoutes()
{
// Custom Routes
RouteTable.Routes.MapUmbracoRoute(
"FDirectory",
"fdirectory/{id}",
new
{
controller = "Directory",
action = "FDirectory",
id = UrlParameter.Optional
},
new PublishedPageRouteHandler(1000));
RouteTable.Routes.MapUmbracoRoute(
"SDirectory",
"sdirectory/{id}",
new
{
controller = "Directory",
action = "SDirectory",
id = UrlParameter.Optional
},
new PublishedPageRouteHandler(1001));
RouteTable.Routes.MapUmbracoRoute(
"HomePage",
"",
new
{
controller = "HomePage",
action = "Index",
id = UrlParameter.Optional
},
new PublishedPageRouteHandler(1002));
}
}
public class PublishedPageRouteHandler : UmbracoVirtualNodeRouteHandler
{
private readonly int _pageId;
public PublishedPageRouteHandler(int pageId)
{
_pageId = pageId;
}
protected override IPublishedContent FindContent(RequestContext requestContext, UmbracoContext umbracoContext)
{
if (umbracoContext != null)
{
umbracoContext = ContextHelpers.EnsureUmbracoContext();
}
var helper = new UmbracoHelper(UmbracoContext.Current);
return helper.TypedContent(_pageId);
}
}
}
DirectoryController.cs
namespace MyApp.Core.Directory.Controllers
{
public class DirectoryController : RenderMvcController
{
public DirectoryController() : this(UmbracoContext.Current) { }
public DirectoryController(UmbracoContext umbracoContext) : base(umbracoContext) { }
public ActionResult FDirectory(RenderModel model)
{
return CurrentTemplate(new DirectoryModel(model.Content));
}
public ActionResult SDirectory(RenderModel model)
{
return CurrentTemplate(new DirectoryModel(model.Content));
}
}
}
So Umbraco does not install with an App_Start folder. I would like to know what the best approach is for a multi-site installation of Umbraco for registering the routes to the controllers. My implementation works, but it seems like I shouldn't have to create actions for every single page I am going to have in a site, in every controller. I know Umbraco has its own routing, so using Umbraco concepts, ASP.NET MVC concepts, and whatever else is available, what is the best way to implement this type of solution structure? Should I even worry about using a RouteConfig.cs and create a App_Start directory? Or is what I am doing the best approach? Should I use IApplicationEventHandler or ApplicationEventHandler?
Also, I have to hard code the node ID's. I've read that there is a way to Dynamically? And example of this would be great.
Examples of the best way to implement a structured multi-site Umbraco MVC solution is what I am asking for I guess, in regards to routing the controllers, with some detail, or links to strong examples. I have searched and researched, and there are bits and pieces out there, but not really a good example like what I am working with. I am going to have to create a RouteMap for every single page I create at this point, and I don't know if this is the most efficient way of doing this. I even tried implementing a DefaultController, but didn't see the point of that when your solution is going to have multiple controllers.
I'm not entirely sure what you are trying to achieve with this, but I'll try to explain how it works and maybe you can clarify afterwards.
I assume you have the basics of Umbraco figured out (creating document types + documents based on the document types). This is how Umbraco is normally used and it will automatically do routing for you for each of these "content nodes" (documents) you create in a site.
So create a document named document1 and it will be automatically routed in your site at URL: http://localhost/document1. By default this document will be served through a default MVC controller and it will all take place behind the scenes without you having to do anything.
Route hijacking allows you to override this default behavior and "shove in" a controller that lets you interfere with how the request is handled. To use hijacking you create a RenderMvcController with the alias of your document type. That could be HomePageController : RenderMvcController.
This controller should have an action with the following signature:
public override ActionResult Index(RenderModel model)
In this action you are able to modify the model being sent to the view in any way you like. That could be - getting some external data to add on to the model or triggering some logic or whatever you need to do.
This is all automatically hooked up by naming convention and you will not have to register any routes manually for this to work.
The other type of Umbraco MVC controller you can create is a SurfaceController. This one is usually used for handling rendering of child actions and form submissions (HttpPost). The SurfaceController is also automatically routed by Umbraco and will be located on a "not so pretty" URL. However since it is usually really not used for anything but rendering child actions and taking form submits, it doesn't really matter what URL it is located at.
Besides these auto-routed controllers you are of course able to register your own MVC controllers like in any standard MVC website. The one difference though is that unlike a normal ASP.NET MVC website, an Umbraco site does not have the automagical default registration of controllers allowing the routing to "just work" when creating a new controller.
So if you want to have a plain old MVC controller render in an Umbraco site without it being related to a document/node in Umbraco, you would have to register a route for it like you would do in any other MVC site. The best way of doing that is to hook in and add it to the Routes using an ApplicationEventHandler class. That will automatically be triggered during application startup - essentially allowing you to do what you would normally do in App_Start.
Just to be clear though - if you plan on using data from Umbraco, you should not be using normal MVC controllers and should not require any manual route registration to be done. You usually want to render a template/view in context of a document/node created in Umbraco (where you can modify data/properties of the document) and then the route hijacking is the way to go.
From what it looks like, it could seem that the correct way to do what you are trying to do is to simply create two document types:
FDirectory and SDirectory
You click to allow both of these to be created in root and then you create documents called FDirectory and SDirectory and they will be automatically routed on these URLs. Creating a RenderMvcController's called FDirectoryController : RenderMvcController will then make sure it is used to hijack the routing whenever that page is requested.
If you're simply trying to set up a multi-site solution I would suggest you create a Website document type and create a node for each site you want, in the root of your Umbraco content tree. Right click each of these nodes and edit the hostname to be whatever you need it to be. This can also be some "child url" like /fdirectory or /sdirectory in case you need to test this on localhost without using multiple hostnames.
Hope this gives you the pointers needed, otherwise try to explain what you are trying to do and I'll see if I can refine my answer a bit!
I'm trying to configure routing in my application in such way: paths to files should work for every directory except one special. And access to files in this direcotory should be processed by my controller action. If I write smth like this:
routes.Map("Specialfolder/{name}", "Controller", "Action");
it works only for unexisting files. Controller doesn't catches route for existing file
If I add this line:
routes.RouteExistingFiles = true;
Working with files in my folder is ok. But files in other directories aren't routed anymore. How to fix this?
Here is a sample ActionFilter sample code snippet that you can use
public class UriActionFilter : System.Web.Mvc.ActionFilterAttribute
{
public override void OnActionExecuting(System.Web.Mvc.ActionExecutingContext filterContext)
{
if (System.Threading.Thread.CurrentPrincipal.Identity.IsAuthenticated)
{
// Sample: somehow identify the user, in case of a custom identity, replace the below line to get the user identifier
var user = System.Threading.Thread.CurrentPrincipal.Identity.Name;
// get the parameter that will let you know what is the image or path that is being requested now
var ctxParams = filterContext.ActionParameters;
// if the user does not have the permission to view, return 403
filterContext.RequestContext.HttpContext.Response.StatusCode = 403;
return;
}
base.OnActionExecuting(filterContext);
}
}
In this snippet, you have to replace with the corresponding service calls etc..
W.R.TO the OnAuthorization, I was mentioning that you can use so that any one that gets access to the image may hit the URI directly and that can be used filtered. We are using that way to restrict the URI access directly by a given user.
This will help you to override the MVC authorization of each request that the controller gets. This is a more of a restriction by URI alone. However, your case can be satisfied by the above action filter too as parsing of the uri parameters can be quite tricky at times.
TL;DR
I need a way to programtically choose which RoutePrefix is chosen when generating URLs based on the properties of a user in my MVC app
Not TL;DR
I have an MVC 4 app (with the AttributeRouting NuGet package)
Due to the requirements of the hosting environment I have to have two routes for a lot of my actions so that the hosting environment can have different permissions for access.
I am solving this by decorating my controller with with [RoutePrefix("full")] [RoutePrefix("lite)]. which allows each action method to be accessed via /full/{action} and /lite/{action}.
This works perfectly.
[RoutePrefix("full")]
[RoutePrefix("lite")]
public class ResultsController : BaseController
{
// Can be accessed via /full/results/your-results and /lite/results/your-results and
[Route("results/your-results")]
public async Task<ActionResult> All()
{
}
}
However, each user should only use either full or lite in their urls, which is determined by some properties of that user.
Obviously when I use RedirectToAction() or #Html.ActionLink() it will just choose the first available route and won't keep the "correct" prefix.
I figured I can override the RedirectToAction() method as well as add my own version of #Html.ActionLink() methods.
This will work, but it will involve some nasty code for me to generate the URLs because all I get is a string representing the action and controllers, but not the reflected types. Also there might be route attributes such as in my example, so I am going to have to replicated a lot of MVCs built in code.
Is there a better solution to the problem I am trying to solve?
How about something like:
[RoutePrefix("{version:regex(^full|lite$)}")]
Then, when you create your links:
#Url.RouteUrl("SomeRoute", new { version = "full" })
Or
#Url.RouteUrl("SomeRoute", new { version = "lite" })
You could even do the following to just keep whatever was already set:
#Url.RouteUrl("SomeRoute", new { version = Request["version"] })
I ended up finding a solution to this
I just overrided the default routes to include this. ASP.Net automatically keeps the usertype value and puts it back in when it regenerates the routes
const string userTypeRegex = "^(full|lite)$";
routes.Add("Default", new Route("{usertype}/{controller}/{action}/{id}",
new { controller = "Sessions", action = "Login", id = UrlParameter.Optional }, new { usertype = userTypeRegex }));
I found that this didn't work with the Route or RoutePrefix attributes, and so I had to remove all of them. Forcing me to add specific routes in these cases
routes.Add("Profile-Simple", new Route("{usertype}/profile/simple",
new { controller = "ProfileSimple", action = "Index" }, new { usertype = userTypeRegex }));
I thought that a half-dozen hard coded routes in my RouteConfig file was a better solution than having to manually add values to each place I generated a URL (as in Chris's solution).
I've come across a bit of a weird problem I can't make sense of. One of my controllers has stopped working, but if I rename it then it works fine. I don't have any special routing wrapped around this controller, it just uses my default.
To give specifics, I have a controller called "Kangaroo". In the browser, if I go to {server}/Kangaroo, then I get the "The Resource cannot be found" message. However, if I go to {server}/Kangaroo/Index, then my page loads as normal. I don't have this problem on any of my other controllers, only this one. If I rename the controller (and my view folder) to "Kangaroo2", then it works perfectly fine.
Here is my route:
public class RouteDefinitions {
public static void AddRoutes(RouteCollection routes) {
routes.Ignore("{resource}.axd/{*pathInfo}");
routes.MapRoute("Resources",
"cache/{action}/{key}/{version}/{type}",
new { controller = "Cache",
action = "CacheContent",
key = "",
version = "",
type = "" });
routes.MapRoute("Default", // Route name
"{controller}/{action}/{id}", // URL with parameters
new {
controller = "Home",
action = "Index",
id = ""
} // Parameter defaults
);
}
}
Does anyone have an idea of what could be going on here? I thought it might just be a weird visual studio thing, but restarting did not correct the issue.
Just figured out what the problem was. There was a folder in my project called "/Kangaroo". I guess it was treating it like a script or other content. Since the path existed it was attempting to load something from the path.
I'd like to create a dynamic thumbnail resizer so that you can use the following URL to get a resized image:
http://server/images/image.jpg?width=320&height=240
I tried setting up a route like this:
routes.MapRoute(null,
"{filename}",
new { controller = "Image", action = "Resize" });
But if the file exists at the URL, ASP.NET will bypass the routing and return you just the file instead. How do I force ASP.NET to route the images instead of returning what's on disk?
Why not just use an action to do this? A controller's action can stream back an image. Otherwise, the typical way, say with ASPX, is that a handler or handler factory listens for the file extension and processes it accordingly. Or use URL rewriting to rewrite the URL in the request.
Thats how asp.net routing works, there is no away around that... you have to use Rewrite if you want to intercept requests for existing files.
Update
Seems like i was a bit too fast on the trigger there. There seems to be a property you can set which allows you to enforce a route even for existing files.
RouteCollection.RouteExistingFiles Property
http://msdn.microsoft.com/en-us/library/system.web.routing.routecollection.routeexistingfiles.aspx
Gets or sets a value that indicates whether ASP.NET routing should handle URLs that match an existing file. True if ASP.NET routing handles all requests, even those that match an existing file; otherwise, false. The default value is false.
You could also consider:
Writing a Module to handle these image routes before it hits routing (registered in Web.Config)
Write your own route handler specifically to handle these images.
Both would allow you to remove the need to write as a controller, I think this is cleaner.
Very basic example of your own route handler (from memory)...
Register as a normal route:
/* Register in routing */
routes.Add("MyImageHandler",
new Route("my-custom-url/{folder}/{filename}",
new ImageRouteHandler())
);
/* Your route handler */
public class ImageRouteHandler : IRouteHandler
{
public IHttpHandler GetHttpHandler(RequestContext requestContext)
{
string filename = requestContext.RouteData.Values["filename"] as string;
string folder = requestContext.RouteData.Values["folder"] as string;
string width = requestContext.HttpContext.Request.Params["w"] as string;
string height = requestContext.HttpContext.Request.Params["h"] as string;
// Look up the file and handle and return, etc...
}
}
Hope these help. Lots of ways to extend and achieve :)
The simplest way would be to route all images through the controller and store your images in a separate location
routes.MapRoute("Images",
"/images/{filename}",
new { controller = "Image", action = "Resize" });
/sitebase/images/image.jpg //public image location
/sitebase/content/images/image.jpg //real image location
Your controller would then see which image was being requested and load the appropriate file from the file system. This would allow you to do what you want without any special handling.
How about:
routes.MapRoute("Images",
"/images/{filename}.jpg",
new { controller = "Image", action = "Resize" });
That Should ensure that only URLs with .jpg as an extension get matched to that route and get routed appropriately.
Also remember you want to add your actions in order of most specific to least specific, with your default one being added last.
Of course, your action still needs to serve up the Image using a filecontentresult.