GetVirtualPath method in mvc - asp.net-mvc

Can anyone explain how the virtual path is being calculating?
According to the RouteData.Values or according to the url pattern?
I'm trying to remove some routedata values but still the virtual path is not changes.
I have a problem that the virtual path return with redundant slash at the beginning of the URL like : /he/controller/action the slash before culture is redundant...
I'm using custom routes like the following
routes.Add("Default",
new CustomRoute("{culture}/{controller}/{action}/{id}",
new
{
controller = "Desktop",
action = "Index",
culture = "he-IL",
guid = "",
id = UrlParameter.Optional
}));
routes.Add("Wizard_" + wizard,
new CustomRoute("{guid}/{culture}/" + wizardName + "/{action}/{id}",
new
{
controller = wizard,
action = "Index",
culture = "he-IL",
guid = "",
id = UrlParameter.Optional
}));
the problem is when using Url.Action(action, controller) method and the action is in the wizard controller, so the URL for the action is wizard format like {guid}/{culture}/" + wizard + "/{action}/{id}
bu the guid value is empty and the returned URL is //he-il/controller/action
instead of /he-il/controller/action
The CustomRoute class:
public class CustomRoute : Route
{
private List<string> _wizards;
public CustomRoute(string uri, object defaults)
: base(uri, new RouteValueDictionary(defaults), new MvcRouteHandler())
{
_wizards = new List<string>();
FillWizards(ref _wizards);
DataTokens = new RouteValueDictionary();
}
public override RouteData GetRouteData(HttpContextBase httpContext)
{
bool hasGuid = httpContext.Request.RequestContext.RouteData != null
&& httpContext.Request.RequestContext.RouteData.Values != null
&& httpContext.Request.RequestContext.RouteData.Values.ContainsKey("guid")
&& !httpContext.Request.RequestContext.RouteData.Values["guid"].ToString().Equals(Guid.Empty);
var routeData = base.GetRouteData(httpContext);
if (routeData == null)
return null;
bool isWizard = _wizards.Contains(routeData.Values["controller"].ToString());
Debug.WriteLine("Controller: " + routeData.Values["controller"] + " action: " + routeData.Values["action"] + " Is wizard: " + isWizard + " has guid: " + hasGuid);
if (isWizard && !hasGuid)
{
if (string.IsNullOrEmpty(routeData.Values["guid"].ToString()))
{
routeData.Values["guid"] = Guid.NewGuid().ToString("N");
}
}
return routeData;
}
public override VirtualPathData GetVirtualPath(RequestContext requestContext, RouteValueDictionary values)
{
VirtualPathData path;
path = base.GetVirtualPath(requestContext, values);
return path;
}
private void FillWizards(ref List<string> items)
{
var _configuration = ObjectFactory.GetInstance<IConfiguration>();
List<string> wizards = _configuration.GetParamValue<string>("SessionUniqueWizards", "").Split(new char[] { ',' }, StringSplitOptions.RemoveEmptyEntries).ToList();
items = wizards;
}
}

The reason why your guid parameter is missing is because
There is no guid parameter in your call to Url.Action(action, controller).
There is (apparently) no guid parameter in the current request. That is, the current route that is being hit has no guid route value.
You have specified the default value for the guid (guid = ""). Since empty string is what you specified as the default, empty string is what you are getting by default.
For the URL to build correctly, the guid has to come from somewhere. MVC always passes matching route values from the current request when building outgoing URLs, but since not all of your URLs have a guid you need to specify it for the pages where it does not exist in the context:
Url.Action("Index", "Search", new { guid = "a565f84f9152495792d433f5bd26000f")
This is the normal way to do it. Typically, if you are building the link for a CRUD operation you are doing so within a list of entities.
For example, for a Product entity, you would normally have links for Edit Product and Delete Product for each Product in a list that would look something like this:
<tr>
<td>model.ProductName</td>
<td>#Html.ActionLink("Product", "Edit", new { guid = model.ProductId })</td>
<td>#Html.ActionLink("Product", "Delete", new { guid = model.ProductId })</td>
</tr>
There would also typically be a link to add a new entity that has no identifier.
#Html.ActionLink("Product", "Add")
But it is unclear from your example how a "wizard" would be created. Creating a new identifier should normally be a function of the Add method, not that of a route.
But there seems to be an issue where you are randomly generating a GUID within your route, so it is unclear how you expect this value to be maintained from one request to the next.

Related

How to customize route to map url dynamically to a composed named controller in ASP.NET MVC3

I need to map URLs like this:
/stock/risk -->StockRiskController.Index()
/stock/risk/attr -->StockRiskController.Attr()
/srock/risk/chart -->StockRiskController.Chart()
...
/bond/performance -->BondPerformanceController.Index()
/bond/performance/attr -->BondPerformanceController.Attr()
/bond/performance/chart -->BondPerformanceController.Chart()
...
The first part is dynamic but enumerable, the second part has only two options(risk|performance).
For now I know only two ways:
customized a ControllerFactory(seems overkilled or complicated)
hard code all the combinations because they are enumerable(ugly).
Can I use routes.MapRoute to achieve this? Or any other handy way?
There is a nice solution based on IRouteConstraint. First of all we have to create new route mapping:
routes.MapRoute(
name: "PrefixedMap",
url: "{prefix}/{body}/{action}/{id}",
defaults: new { prefix = string.Empty, body = string.Empty
, action = "Index", id = string.Empty },
constraints: new { lang = new MyRouteConstraint() }
);
Next step is to create our Constraint. Before I will introduce some way how to check relevance as mentioned above - two list with possible values, but logic could be adjusted
public class MyRouteConstraint : IRouteConstraint
{
public readonly IList<string> ControllerPrefixes = new List<string> { "stock", "bond" };
public readonly IList<string> ControllerBodies = new List<string> { "risk", "performance" };
...
And now the Match method, which will adjust the routing as we need
public bool Match(System.Web.HttpContextBase httpContext
, Route route, string parameterName, RouteValueDictionary values
, RouteDirection routeDirection)
{
// for now skip the Url generation
if (routeDirection.Equals(RouteDirection.UrlGeneration))
{
return false;
}
// try to find out our parameters
string prefix = values["prefix"].ToString();
string body = values["body"].ToString();
var arePartsKnown =
ControllerPrefixes.Contains(prefix, StringComparer.InvariantCultureIgnoreCase) &&
ControllerBodies.Contains(body, StringComparer.InvariantCultureIgnoreCase);
// not our case
if (!arePartsKnown)
{
return false;
}
// change controller value
values["controller"] = prefix + body;
values.Remove("prefix");
values.Remove("body");
return true;
}
You can play with this method more, but the concept should be clear now.
NOTE: I like your approach. Sometimes it is simply much more important to extend/adjust routing then go to code and "fix names". Similar solution was working here: Dynamically modify RouteValueDictionary

MVC RouteDatas are confused

I am using asp.net mvc for my website project. I think i have wrong things in my routedata but i am not sure it is wrong or ok. i will explain the situation.
I am caching my action results (html outputs) in Cache with a generated key
public static string GetKeyFromActionExecutingContext(ControllerContext filterContext)
{
StringBuilder keyBuilder = new StringBuilder();
if (filterContext.IsChildAction)
keyBuilder.Append("C-");
else
keyBuilder.Append("P-");
foreach (var item in filterContext.RouteData.Values)
{
keyBuilder.AppendFormat("{0}={1}.", item.Key, item.Value);
}
return keyBuilder.ToString();
}
ex: For HomePage , generated cache key is P-Controller=Home.Action=Index and
I have also childactions in my sitemaster like LoginBox(It is in MembershipController/LoginBox)
Its cache key is C-Controller=Membership.Action=LoginBox.
Everything is okey till now.
I have also subcategories in my website like
domain/category1
domain/category1/subcategory1
domain/category1/subcategory2
domain/category2
When i am browsing a sub category from domain/category1
My generated keys are failed because my routedatas are wrong
filterContext.RouteData.Values:
Controller = Membership
Action = LoginBox
ctg1 = category1
ctg2 = ""
ctg3 = ""
Why these are mixed. It is using the "Category" routemapping but I think it must use "Default" routemapping.
My global.asax like below
routes.MapRoute(
"Category",
"{ctg0}/{ctg1}/{ctg2}/{ctg3}",
new
{
controller = "Category",
action = "Index",
ctg0 = "",
ctg1 = "",
ctg2 = "",
ctg3 = ""
},
new
{
ctg0 = new CategoryRouteConstraint(),
}
);
routes.MapRoute(
"Default",
"{controller}/{action}/{id}", new { controller = "Home", action = "Index", id = "" },
new { controller = #"[^\.]*" }
);
Also my CategoryRouteConstraint Method it is checking from db that ctg0 value is a category name
public class CategoryRouteConstraint : IRouteConstraint
{
public Boolean Match(
HttpContextBase httpContext,
Route route,
String sParameterName,
RouteValueDictionary values,
RouteDirection routeDirection
)
{
if ((routeDirection == RouteDirection.IncomingRequest))
{
if (values["ctg0"] != null && !string.IsNullOrEmpty(values["ctg0"].ToString()))
return Category.IsRoutingForCategory(values["ctg0"].ToString());
return false;
}
return false;
}
}
Hopefully this may help you, it will show you which routes a url matches, I was a little confused by the question.
http://haacked.com/archive/2008/03/13/url-routing-debugger.aspx

Ambient values in mvc2.net routing

I have following two routes registered in my global.asax file
routes.MapRoute(
"strict",
"{controller}.mvc/{docid}/{action}/{id}",
new { action = "Index", id = "", docid = "" },
new { docid = #"\d+"}
);
routes.MapRoute(
"default",
"{controller}.mvc/{action}/{id}",
new { action = "Index", id = "" },
new { docConstraint = new DocumentConstraint() }
);
and I have a static "dashboard" link in my tabstrip and some other links that are constructed from values in db here is the code
<ul id="globalnav" class = "t-reset t-tabstrip-items">
<li class="bar" id = "dashboard">
<%=Html.ActionLink("dash.board", "Index", pck.Controller, new{docid =string.Empty,id = pck.PkgID }, new { #class = "here" })%>
</li>
<%
foreach (var md in pck.sysModules)
{
%>
<li class="<%=liClass%>">
<%=Html.ActionLink(md.ModuleName, md.ActionName, pck.Controller, new { docid = md.DocumentID}, new { #class = cls })%>
</li>
<%
}
%>
</ul>
Now my launching address is localhost/oa.mvc/index/11 clearly matching the 2nd route. But when I visit any page that has mapped to first route and then come back to dash.board link it shows me localhost/oa.mvc/7/index/11 where 7 is docid and picked from previous Url.
I understand that my action method is after docid and changing it would not clear the docid.
My question here is, can I remove docid in this scenario without changing the route?
I have the same "not clearing out" value problem...
I've stepped into source code and I don't understand the reason for being of segment commented as : // Add all current values that aren't in the URL at all
# System\Web\Routing\ParsedRoute.cs, public BoundUrl Bind(RouteValueDictionary currentValues, RouteValueDictionary values, RouteValueDictionary defaultValues, RouteValueDictionary constraints) method from line 91 to line 100
While the clearing process is correctly handled in method preceding steps, this code "reinjects" the undesired parameter into acceptedValues dictionary!?
My routing is defined this way:
routes.MapRoute(
"Planning",
"Plans/{plan}/{controller}/{action}/{identifier}",
new { controller = "General", action = "Planning", identifier = UrlParameter.Optional },
new { plan = #"^\d+$" }
);
// default application route
routes.MapRoute(
"Default",
"{controller}/{action}/{identifier}",
new {
controller = "General",
action = "Summary",
identifier = UrlParameter.Optional,
plan = string.Empty // mind this default !!!
}
);
This is very similar to what you're using. But mind my default route where I define defaults. Even though my default route doesn't define plan route value I still set it to string.Empty. So whenever I use Html.ActionLink() or Url.Action() and I want plan to be removed from the URL I call it the usual way:
Url.Action("Action", "Controller", new { plan = string.Empty });
And plan is not included in the URL query string any more. Try it out yourself it may work as well.
Muhammad, I suggest something like this :
(written 5 mn ago, not tested in production)
public static class MyHtmlHelperExtensions {
public static MvcHtmlString FixActionLink(this HtmlHelper htmlHelper, string linkText, string actionName, string controllerName, object routeValues, object htmlAttributes) {
var linkRvd = new RouteValueDictionary(routeValues);
var contextRvd = htmlHelper.ViewContext.RouteData.Values;
var contextRemovedRvd = new RouteValueDictionary();
// remove clearing route values from current context
foreach (var rv in linkRvd) {
if (string.IsNullOrEmpty((string)rv.Value) && contextRvd.ContainsKey(rv.Key)) {
contextRemovedRvd.Add(rv.Key, contextRvd[rv.Key]);
contextRvd.Remove(rv.Key);
}
}
// call ActionLink with modified context
var htmlString = htmlHelper.ActionLink(linkText, actionName, controllerName, routeValues, htmlAttributes);
// restore context
foreach (var rv in contextRemovedRvd) {
contextRvd.Add(rv.Key, rv.Value);
}
return htmlString;
}
}
This is such a frustrating problem and I would venture to say that it is even a bug in ASP.Net MVC. Luckily it's an easy fix using ActionFilters. If you are using MVC3 then I would just put this as a global attribute to clear out ambient values. I made this attribute discriminatory, but you can change it to clear all attributes.
The assumption here is that by the time the Result is executing (your view most likely), you have already explicitly specified all your ActionLinks and Form Actions. Thus this will execute before they (the links) are evaluated, giving you a new foundation to generate them.
public class ClearAmbientRouteValuesAttribute : ActionFilterAttribute
{
private readonly string[] _keys;
public ClearAmbientRouteValuesAttribute(params string [] keys)
{
if (keys == null)
_keys = new string[0];
_keys = keys;
}
public override void OnResultExecuting(ResultExecutingContext filterContext)
{
foreach (var key in _keys) {
// Why are you sticking around!!!
filterContext.RequestContext.RouteData.Values.Remove(key);
}
}
}
// Inside your Global.asax
public static void RegisterGlobalFilters(GlobalFilterCollection filters)
{
filters.Add(new HandleErrorAttribute());
filters.Add(new ClearAmbientRouteValuesAttribute("format"));
}
Hope this helps someone, cause it sure helped me. Thanks for asking this question.
In this particular scenario I have two recommendations:
Use named routes. The first parameter to the MapRoute method is a name. To generate links use Html.RouteLink() (and other similar APIs). This way you'll always choose the exact route that you want and never have to wonder what gets chosen.
If you still want to use Html.ActionLink() then explicitly set docid="" to clear out its value.
Here's how I solved my problem, it may take a little adapting to get it to work, but I felt like I could get what I needed and just use routing more or less normally:
Excerpted from Apress Pro ASP.Net.MVC 3 Framework:
A value must be available for every segment variable defined in the URL pattern.
To find values for each segment variable, the routing system looks first at the
values we have provided (using the properties of anonymous type), then the
variable values for the current request, and finally at the default values defined in
the route. (We return to the second source of these values later in this chapter.)
None of the values we provided for the segment variables may disagree with the
default-only variables defined in the route. These are variables for which default
values have been provided, but which do not occur in the URL pattern. For
example, in this route definition, myVar is a default-only variable:
routes.MapRoute("MyRoute", "{controller}/{action}",
new { myVar = "true" });
For this route to be a match, we must take care to not supply a value for myVar or to make
sure that the value we do supply matches the default value.
The values for all of the segment variables must satisfy the route constraints. See
the “Constraining Routes” section earlier in the chapter for examples of different
kinds of constraints.
Basically I used the rule about a route not matching if it doesn't define a segment, but has a default variable used to give me a little more control over whether a route was chosen for outbound routing or not.
Here's my fixed routes, notice how I specify a value for category that would never be valid and don't specify a segment for category. This means that route will be skipped if I have a category, but will use it if I only have a page:
routes.MapRoute(null, "receptionists/faq/{page}", new { controller = "Receptionist", action = "Faq", page = 1, category = (Object)null }, new { page = #"^\d+$" });
routes.MapRoute(null, "receptionists/faq/{category}/{page}", new { controller = "Receptionist", action = "Faq", page = 1 }, new { category = #"^\D+$", page = #"^\d+$" });
For Category Links
#Html.ActionLink("All", "Faq", new { page = 1 })
#foreach (var category in Model.Categories)
{
#Html.ActionLink(category.DisplayName, "faq", new { category = category.DisplayName.ToLower(), page = 1 })
}
For Page Links
#for (var p = 1; p <= Model.TotalPages; p++)
{
#Html.ActionLink(p.ToString(), "Faq", new { page = p, category = Model.CurrentCategory})
}

ASP.NET MVC: Route to URL

What's the easiest way to get the URL (relative or absolute) to a Route in MVC? I saw this code here on SO but it seems a little verbose and doesn't enumerate the RouteTable.
Example:
List<string> urlList = new List<string>();
urlList.Add(GetUrl(new { controller = "Help", action = "Edit" }));
urlList.Add(GetUrl(new { controller = "Help", action = "Create" }));
urlList.Add(GetUrl(new { controller = "About", action = "Company" }));
urlList.Add(GetUrl(new { controller = "About", action = "Management" }));
With:
protected string GetUrl(object routeValues)
{
RouteValueDictionary values = new RouteValueDictionary(routeValues);
RequestContext context = new RequestContext(HttpContext, RouteData);
string url = RouteTable.Routes.GetVirtualPath(context, values).VirtualPath;
return new Uri(Request.Url, url).AbsoluteUri;
}
What's a better way to examine the RouteTable and get a URL for a given controller and action?
Use the UrlHelper class: http://msdn.microsoft.com/en-us/library/system.web.mvc.urlhelper.aspx
You should be able to use it via the Url object in your controller. To map to an action, use the Action method: Url.Action("actionName","controllerName");.
A full list of overloads for the Action method is here: http://msdn.microsoft.com/en-us/library/system.web.mvc.urlhelper.action.aspx
so your code would look like this:
List<string> urlList = new List<string>();
urlList.Add(Url.Action("Edit", "Help"));
urlList.Add(Url.Action("Create", "Help"));
urlList.Add(Url.Action("Company", "About"));
urlList.Add(Url.Action("Management", "About"));
EDIT: It seems, from your new answer, that your trying to build a sitemap.
Have a look at this Codeplex project: http://mvcsitemap.codeplex.com/. I haven't used it myself, but it looks pretty solid.
How about this (in the controller):
public IEnumerable<SiteMapEntry> SiteMapEntries
{
get
{
var entries = new List<SiteMapEntry>();
foreach (var route in this.Routes)
{
entries.Add(new SiteMapEntry
(
this.Url.RouteUrl(route.Defaults),
SiteMapEntry.ChangeFrequency.Weekly,
DateTime.Now,
1F));
}
return entries;
}
}
Where the controller has member:
public IEnumerable<Route> Routes
Take note of:
this.Url.RouteUrl(route.Defaults)

Asp.net MVC ModelState.Clear

Can anyone give me a succinct definition of the role of ModelState in Asp.net MVC (or a link to one). In particular I need to know in what situations it is necessary or desirable to call ModelState.Clear().
Bit open ended huh... sorry, I think it might help if tell you what I'm acutally doing:
I have an Action of Edit on a Controller called "Page". When I first see the form to change the Page's details everything loads up fine (binding to a "MyCmsPage" object). Then I click a button that generates a value for one of the MyCmsPage object's fields (MyCmsPage.SeoTitle). It generates fine and updates the object and I then return the action result with the newly modified page object and expect the relevant textbox (rendered using <%= Html.TextBox("seoTitle", page.SeoTitle)%>) to be updated ... but alas it displays the value from the old model that was loaded.
I've worked around it by using ModelState.Clear() but I need to know why / how it has worked so I'm not just doing it blindly.
PageController:
[AcceptVerbs("POST")]
public ActionResult Edit(MyCmsPage page, string submitButton)
{
// add the seoTitle to the current page object
page.GenerateSeoTitle();
// why must I do this?
ModelState.Clear();
// return the modified page object
return View(page);
}
Aspx:
<%# Page Language="C#" MasterPageFile="~/Views/Shared/Site.Master" Inherits="System.Web.Mvc.ViewPage<MyCmsPage>" %>
....
<div class="c">
<label for="seoTitle">
Seo Title</label>
<%= Html.TextBox("seoTitle", page.SeoTitle)%>
<input type="submit" value="Generate Seo Title" name="submitButton" />
</div>
I think is a bug in MVC. I struggled with this issue for hours today.
Given this:
public ViewResult SomeAction(SomeModel model)
{
model.SomeString = "some value";
return View(model);
}
The view renders with the original model, ignoring the changes. So I thought, maybe it does not like me using the same model, so I tried like this:
public ViewResult SomeAction(SomeModel model)
{
var newModel = new SomeModel { SomeString = "some value" };
return View(newModel);
}
And still the view renders with the original model. What's odd is, when I put a breakpoint in the view and examine the model, it has the changed value. But the response stream has the old values.
Eventually I discovered the same work around that you did:
public ViewResult SomeAction(SomeModel model)
{
var newModel = new SomeModel { SomeString = "some value" };
ModelState.Clear();
return View(newModel);
}
Works as expected.
I don't think this is a "feature," is it?
Update:
This is not a bug.
Please stop returning View() from a POST action. Use PRG instead and redirect to a GET if the action is a success.
If you are returning a View() from a POST action, do it for form validation, and do it the way MVC is designed using the built in helpers. If you do it this way then you shouldn't need to use .Clear()
If you're using this action to return ajax for a SPA, use a web api controller and forget about ModelState since you shouldn't be using it anyway.
Old answer:
ModelState in MVC is used primarily to describe the state of a model object largely with relation to whether that object is valid or not. This tutorial should explain a lot.
Generally you should not need to clear the ModelState as it is maintained by the MVC engine for you. Clearing it manually might cause undesired results when trying to adhere to MVC validation best practises.
It seems that you are trying to set a default value for the title. This should be done when the model object is instantiated (domain layer somewhere or in the object itself - parameterless ctor), on the get action such that it goes down to the page the 1st time or completely on the client (via ajax or something) so that it appears as if the user entered it and it comes back with the posted forms collection. Some how your approach of adding this value on the receiving of a forms collection (in the POST action // Edit) is causing this bizarre behaviour that might result in a .Clear() appearing to work for you. Trust me - you don't want to be using the clear. Try one of the other ideas.
If you want to clear a value for an individual field then I found the following technique useful.
ModelState.SetModelValue("Key", new ValueProviderResult(null, string.Empty, CultureInfo.InvariantCulture));
Note:
Change "Key" to the name of the field that you want to reset.
Well lots of us seem to have been bitten by this, and although the reason this happens makes sense I needed a way to ensure that the value on my Model was shown, and not ModelState.
Some have suggested ModelState.Remove(string key), but it's not obvious what key should be, especially for nested models. Here are a couple methods I came up with to assist with this.
The RemoveStateFor method will take a ModelStateDictionary, a Model, and an expression for the desired property, and remove it. HiddenForModel can be used in your View to create a hidden input field using only the value from the Model, by first removing its ModelState entry. (This could easily be expanded for the other helper extension methods).
/// <summary>
/// Returns a hidden input field for the specified property. The corresponding value will first be removed from
/// the ModelState to ensure that the current Model value is shown.
/// </summary>
public static MvcHtmlString HiddenForModel<TModel, TProperty>(this HtmlHelper<TModel> helper,
Expression<Func<TModel, TProperty>> expression)
{
RemoveStateFor(helper.ViewData.ModelState, helper.ViewData.Model, expression);
return helper.HiddenFor(expression);
}
/// <summary>
/// Removes the ModelState entry corresponding to the specified property on the model. Call this when changing
/// Model values on the server after a postback, to prevent ModelState entries from taking precedence.
/// </summary>
public static void RemoveStateFor<TModel, TProperty>(this ModelStateDictionary modelState, TModel model,
Expression<Func<TModel, TProperty>> expression)
{
var key = ExpressionHelper.GetExpressionText(expression);
modelState.Remove(key);
}
Call from a controller like this:
ModelState.RemoveStateFor(model, m => m.MySubProperty.MySubValue);
or from a view like this:
#Html.HiddenForModel(m => m.MySubProperty.MySubValue)
It uses System.Web.Mvc.ExpressionHelper to get the name of the ModelState property.
Well the ModelState basically holds the current State of the model in terms of validation, it holds
ModelErrorCollection: Represent the errors when the model try to bind the values.
ex.
TryUpdateModel();
UpdateModel();
or like a parameter in the ActionResult
public ActionResult Create(Person person)
ValueProviderResult: Hold the details about the attempted bind to the model.
ex. AttemptedValue, Culture, RawValue.
Clear() method must be use with caution because it can lead to unspected results. And you will lose some nice properties of the ModelState like AttemptedValue, this is used by MVC in the background to repopulate the form values in case of error.
ModelState["a"].Value.AttemptedValue
I had an instance where I wanted to update the model of a sumitted form, and did not want to 'Redirect To Action' for performanace reason. Previous values of hidden fields were being retained on my updated model - causing allsorts of issues!.
A few lines of code soon identified the elements within ModelState that I wanted to remove (after validation), so the new values were used in the form:-
while (ModelState.FirstOrDefault(ms => ms.Key.ToString().StartsWith("SearchResult")).Value != null)
{
ModelState.Remove(ModelState.FirstOrDefault(ms => ms.Key.ToString().StartsWith("SearchResult")));
}
I wanted to update or reset a value if it didn't quite validate, and ran into this problem.
The easy answer, ModelState.Remove, is.. problematic.. because if you are using helpers you don't really know the name (unless you stick by the naming convention). Unless perhaps you create a function that both your custom helper and your controller can use to get a name.
This feature should have been implemented as an option on the helper, where by default is does not do this, but if you wanted the unaccepted input to redisplay you could just say so.
But at least I understand the issue now ;).
Got it in the end. My Custom ModelBinder which was not being registered and does this :
var mymsPage = new MyCmsPage();
NameValueCollection frm = controllerContext.HttpContext.Request.Form;
myCmsPage.SeoTitle = (!String.IsNullOrEmpty(frm["seoTitle"])) ? frm["seoTitle"] : null;
So something that the default model binding was doing must have been causing the problem. Not sure what, but my problem is at least fixed now that my custom model binder is being registered.
Generally, when you find yourself fighting against a framework standard practices, it is time to reconsider your approach. In this case, the behavior of ModelState. For instance, when you don't want model state after a POST, consider a redirect to the get.
[HttpPost]
public ActionResult Edit(MyCmsPage page, string submitButton)
{
if (ModelState.IsValid) {
SomeRepository.SaveChanges(page);
return RedirectToAction("GenerateSeoTitle",new { page.Id });
}
return View(page);
}
public ActionResult GenerateSeoTitle(int id) {
var page = SomeRepository.Find(id);
page.GenerateSeoTitle();
return View("Edit",page);
}
EDITED to answer culture comment:
Here is what I use to handle a multi-cultural MVC application. First the route handler subclasses:
public class SingleCultureMvcRouteHandler : MvcRouteHandler {
protected override IHttpHandler GetHttpHandler(RequestContext requestContext)
{
var culture = requestContext.RouteData.Values["culture"].ToString();
if (string.IsNullOrWhiteSpace(culture))
{
culture = "en";
}
var ci = new CultureInfo(culture);
Thread.CurrentThread.CurrentUICulture = ci;
Thread.CurrentThread.CurrentCulture = CultureInfo.CreateSpecificCulture(ci.Name);
return base.GetHttpHandler(requestContext);
}
}
public class MultiCultureMvcRouteHandler : MvcRouteHandler
{
protected override IHttpHandler GetHttpHandler(RequestContext requestContext)
{
var culture = requestContext.RouteData.Values["culture"].ToString();
if (string.IsNullOrWhiteSpace(culture))
{
culture = "en";
}
var ci = new CultureInfo(culture);
Thread.CurrentThread.CurrentUICulture = ci;
Thread.CurrentThread.CurrentCulture = CultureInfo.CreateSpecificCulture(ci.Name);
return base.GetHttpHandler(requestContext);
}
}
public class CultureConstraint : IRouteConstraint
{
private string[] _values;
public CultureConstraint(params string[] values)
{
this._values = values;
}
public bool Match(HttpContextBase httpContext,Route route,string parameterName,
RouteValueDictionary values, RouteDirection routeDirection)
{
// Get the value called "parameterName" from the
// RouteValueDictionary called "value"
string value = values[parameterName].ToString();
// Return true is the list of allowed values contains
// this value.
return _values.Contains(value);
}
}
public enum Culture
{
es = 2,
en = 1
}
And here is how I wire up the routes. After creating the routes, I prepend my subagent (example.com/subagent1, example.com/subagent2, etc) then the culture code. If all you need is the culture, simply remove the subagent from the route handlers and routes.
public static void RegisterRoutes(RouteCollection routes)
{
routes.IgnoreRoute("{resource}.axd/{*pathInfo}");
routes.IgnoreRoute("Content/{*pathInfo}");
routes.IgnoreRoute("Cache/{*pathInfo}");
routes.IgnoreRoute("Scripts/{pathInfo}.js");
routes.IgnoreRoute("favicon.ico");
routes.IgnoreRoute("apple-touch-icon.png");
routes.IgnoreRoute("apple-touch-icon-precomposed.png");
/* Dynamically generated robots.txt */
routes.MapRoute(
"Robots.txt", "robots.txt",
new { controller = "Robots", action = "Index", id = UrlParameter.Optional }
);
routes.MapRoute(
"Sitemap", // Route name
"{subagent}/sitemap.xml", // URL with parameters
new { subagent = "aq", controller = "Default", action = "Sitemap"}, new[] { "aq3.Controllers" } // Parameter defaults
);
routes.MapRoute(
"Rss Feed", // Route name
"{subagent}/rss", // URL with parameters
new { subagent = "aq", controller = "Default", action = "RSS"}, new[] { "aq3.Controllers" } // Parameter defaults
);
/* remap wordpress tags to mvc blog posts */
routes.MapRoute(
"Tag", "tag/{title}",
new { subagent = "aq", controller = "Default", action = "ThreeOhOne", id = UrlParameter.Optional}, new[] { "aq3.Controllers" }
).RouteHandler = new MultiCultureMvcRouteHandler(); ;
routes.MapRoute(
"Custom Errors", "Error/{*errorType}",
new { controller = "Error", action = "Index", id = UrlParameter.Optional}, new[] { "aq3.Controllers" }
);
/* dynamic images not loaded from content folder */
routes.MapRoute(
"Stock Images",
"{subagent}/Images/{*filename}",
new { subagent = "aq", controller = "Image", action = "Show", id = UrlParameter.Optional, culture = "en"}, new[] { "aq3.Controllers" }
);
/* localized routes follow */
routes.MapRoute(
"Localized Images",
"Images/{*filename}",
new { subagent = "aq", controller = "Image", action = "Show", id = UrlParameter.Optional}, new[] { "aq3.Controllers" }
).RouteHandler = new MultiCultureMvcRouteHandler();
routes.MapRoute(
"Blog Posts",
"Blog/{*postname}",
new { subagent = "aq", controller = "Blog", action = "Index", id = UrlParameter.Optional}, new[] { "aq3.Controllers" }
).RouteHandler = new MultiCultureMvcRouteHandler();
routes.MapRoute(
"Office Posts",
"Office/{*address}",
new { subagent = "aq", controller = "Offices", action = "Address", id = UrlParameter.Optional }, new[] { "aq3.Controllers" }
).RouteHandler = new MultiCultureMvcRouteHandler();
routes.MapRoute(
"Default", // Route name
"{controller}/{action}/{id}", // URL with parameters
new { subagent = "aq", controller = "Home", action = "Index", id = UrlParameter.Optional }, new[] { "aq3.Controllers" } // Parameter defaults
).RouteHandler = new MultiCultureMvcRouteHandler();
foreach (System.Web.Routing.Route r in routes)
{
if (r.RouteHandler is MultiCultureMvcRouteHandler)
{
r.Url = "{subagent}/{culture}/" + r.Url;
//Adding default culture
if (r.Defaults == null)
{
r.Defaults = new RouteValueDictionary();
}
r.Defaults.Add("culture", Culture.en.ToString());
//Adding constraint for culture param
if (r.Constraints == null)
{
r.Constraints = new RouteValueDictionary();
}
r.Constraints.Add("culture", new CultureConstraint(Culture.en.ToString(), Culture.es.ToString()));
}
}
}
Well, this seemed to work on my Razor Page and never even did a round trip to the .cs file.
This is old html way. It might be useful.
<input type="reset" value="Reset">

Resources