UPDATE
My original assumption was that optional parameters were the cause of the problem. That appears to be incorrect. Instead, it appears to be a problem with multiple action methods when one of those methods contains nullable value types (e.g. int? ) for some of the parameters.
I'm using Visual Studio 2012 RC and am just getting started with Web API. I've run into an issue and getting the error "No action was found on the controller 'Bars' that matches the request."
I've got a Bars controller. It has a Get() method that takes in optional parameters.
public IEnumerable<string> Get(string h, string w = "defaultWorld", int? z=null)
{
if (z != 0)
return new string[] { h, w, "this is z: " + z.ToString() };
else
return new string[] { h, w };
}
So, I test it out with the following urls
/api/bars?h=hello
/api/bars?h=hello&w=world
/api/bars?h=hello&w=world&z=15
And it works for all three.
Then, I go to add another Get() method, this time with a single id parameter
public string Get(int id)
{
return "value";
}
I test the urls again. This time /api/bars?h=hello&w=world and api/bars?h=hello fail. The error message is "No action was found on the controller 'Bar' that matches the request."
For some reason, these two methods don't play nicely together. If I remove Get(int id), it works. If I change int? z to string z, then it works (, but then it requires converting the objects inside my action method!).
Why is Web API doing this? Is this a bug or by design?
Many thanks.
I haven't found a true answer for this issue yet (why is Web API doing this), but I have a workaround that does allow for an overloaded Get(). The trick is to wrap the parameter values in an object.
public class Foo
{
public string H { get; set; }
public string W { get; set; }
public int? Z { get; set; }
}
And to the Bars controller modify to
public IEnumerable<string> Get([FromUri] Foo foo)
{
if (foo.Z.HasValue)
return new string[] { foo.H, foo.W, "this is z: " + foo.Z.ToString() };
else
return new string[] { foo.H, foo.W, "z does not have a value" };
}
[FromUri] is necessary, because WebAPI does not, by default, use URI parameters to form "complex" objects. The general thinking is that complex objects are coming from <form> actions, not GET requests.
I'm still going keep checking about why Web API behaves this way and if this is actually a bug or intended behavior.
Problem solved, although, it leaves an additional question. The problem appears to be that the overloaded Action methods are having problems with the optional parameters.
So the new question is why so, but I will leave that up to lower level guys than me ;)
But this is good news. I didn't like the problem you reported, and going the complex type route, while nice to know, is simply a jerry rig fix and would reflect very poorly on how something is working in the Web Api. So the good news is, if you have this problem, it is solved by simply doing away with the optional params, do the good ol' overloads route. Good news, as this is by no means a jerry rig fix, simply makes you loose a little optional parameter convenience:
public class BarsController : ApiController
{
public string Get(int id)
{
return "value";
}
public IEnumerable<string> Get(string h)
{
return Get(h, null, null);
}
public IEnumerable<string> Get(string h, string w)
{
return Get(h, w, null);
}
public IEnumerable<string> Get(string h, string w, int? z)
{
if (z != 0)
return new string[] { h, w, "this is z: " + z.ToString() };
else
return new string[] { h, w };
}
}
Cheers
You can overload WEB API Controller methods by adding an action parameter in the route.
routes.MapHttpRoute(
name: "DefaultApi",
routeTemplate: "api/{controller}/{action}/{id}",
defaults: new {action = "Get", id = RouteParameter.Optional }
);
Once you make this change in your route then you can call your methods like
/api/bars/get?id=1
/api/bars/get?h=hello&w=world&z=15
Hope this help.
Omar
Related
I have written the below simple Web API method.
[HttpGet]
[HttpPost]
public int SumNumbers([FromUri]Numbers calc, [FromUri]Operation op)
{
int result = op.Add ? calc.First + calc.Second : calc.First - calc.Second;
return op.Double ? result * 2 : result;
}
Below is the model class for Numbers:
public class Numbers
{
public int First { get; set; }
public int Second { get; set; }
}
Below is the model class for Operation:
public class Operation
{
public bool Add { get; set; }
public bool Double { get; set; }
}
Below is how I am trying to test in Postman. But, as you can see I am getting "0" as output. When debugged the code, understood that values are not passing from Postman into code.
One another user also posted the same problem here. But, whatever the resolution he showed, I am doing already, but I am not getting answer.
Can anyone please suggest where I am doing wrong?
There are 2 major issues with your post, firstly your controller (due to [FromUri]
binding) is specifying that the arguments need to be passed as Query String parameters and not Http Header values.
The second issue is that you have defined two complex type parameters that you want to obtain the values form the URI, this is not supported.
How to pass in Uri complex objects without using custom ModelBinders or any serialization?
This is a great writeup on how to fully exploit the [FromUriAttribute][2] up to ASP.Net Core 2.2, many of the principals apply to the FromQueryAttribute which is still used the current in ASP.Net 6.
We can use [FromUri] to bind multiple primitive typed parameters, or we can bind 1 single complex typed parameter. You cannot combine the two concepts, the reason for this is that when a complex type is used ALL of the query string arguments are bound to that single complex type.
So your options are to create a new complex type that combines all the properties from both types, or declare all the properties of both types as primitive parameters to the method:
http://localhost:29844/api/bindings/SumNumbers1?First=3&Second=2&Add=True&Double=False
http://localhost:29844/api/bindings/SumNumbers2?First=3&Second=2&Add=True&Double=False
[HttpGet]
[HttpPost]
public int SumNumbers1([FromUri] int First, [FromUri] int Second, [FromUri] bool Add, [FromUri] bool Double)
{
int result = Add ? First + Second : First - Second;
return Double ? result * 2 : result;
}
[HttpGet]
[HttpPost]
public int SumNumbers2([FromUri] SumRequest req)
{
int result = req.Add ? req.First + req.Second : req.First - req.Second;
return req.Double ? result * 2 : result;
}
public class SumRequest
{
public int First { get; set; }
public int Second { get; set; }
public bool Add { get; set; }
public bool Double { get; set; }
}
It is also technically possible to use a nested structure, where you wrap the multiple complex with a single outer complex type. Depending on your implementation and host constraints you may need additional configuration to support using . in the query parameters, but a nested or wrapped request implementation would look like this:
http://localhost:29844/api/bindings/SumNumbers3?Calc.First=3&Calc.Second=2&Op.Add=True&Op.Double=False
[HttpGet]
[HttpPost]
public int SumNumbers3([FromUri] WrappedRequest req)
{
int result = req.Op.Add ? req.Calc.First + req.Calc.Second : req.Calc.First - req.Calc.Second;
return req.Op.Double ? result * 2 : result;
}
public class WrappedRequest
{
public Numbers Calc { get; set; }
public Operation Op { get; set; }
}
It is also possible to use a combination of Http Headers and query string parameters, however these are generally harder (less common) to manage from a client perspective.
It is more common with complex parameter scenarios (not to mention more REST compliant) to force the caller to use POST to access your calculation endpoint, then multiple complex types are simpler to support from both a client and API perspective.
If you want to receive parameters using FromUri, shouldn't you pass them in the URL when doing the GET call? A simpler call would be something like this:
[HttpGet]
[Route("{first:int}/{second:int}")]
public int SumNumbers([FromUri]int first, [FromUri]int second)
{
return first+second;
}
And in Postman your call should be more like this (the url)
http://localhost:29844/api/bindings/SumNumbers/5/7
and this would return 12!
Now if you want to pass First and Second as headers and not in the url then you don't want to use FromUri and then your code would change a little bit (you will then need to read the request and dissect it to get every header alone. Something like this:
HttpRequestMessage request = Request ?? new HttpRequestMessage();
string first = request.Headers.GetValues("First").FirstOrDefault();
string second = request.Headers.GetValues("Second").FirstOrDefault();
I'm following Scott Allen's MVC4 course on PluralSight (I'm using MVC5 and WebAPI2 but they should be the same) and I am trying to pass an object via HTTP PUT. The model binder should bind it, but I am getting NULL for the parameter.
public HttpResponseMessage PutObjective(int id, [FromBody] Objective objective)
{
if (ModelState.IsValid && id == objective.ObjectiveID)
{
//todo: update - look up id, replace text
return Request.CreateResponse(HttpStatusCode.OK, objective);
}
else
{
return Request.CreateResponse(HttpStatusCode.BadRequest);
}
}
and in my front-end javascript I am doing the following (I'm creating an object for testing, so ignore 'objective' passed in):
var updateObjective = function (objective) {
var myobj = { "ObjectiveID": "3", "ObjectiveDescription": "test" };
return $.ajax(objectiveApiUrl + "/" + objective.ObjectiveID, {
type: "PUT",
data: myobj
});
}
My class looks like this:
public class Objective
{
public int ObjectiveID { get; private set; }
public string ObjectiveDescription { get; set; }
public Objective (int Id, string Desc)
{
this.ObjectiveID = Id;
this.ObjectiveDescription = Desc;
}
}
Any thoughts on why 'objective' in the backend is always 'null' ?
I've done what Scott Allen is doing, even tried adding in [FromBody] but no luck. $.ajax should have the correct content type by default I understand, so no need to set it.
I had Fiddler2 but I'm unsure as to what I am looking at to be honest. I can see my object as JSON being sent to the backend.
Well, if you're familiar with Model Binding you'll have seen the issue in my Objective class:
public int ObjectiveID { get; private set; }
with a private set, no instance can be created of the Objective class. To make it work, the 'private' access specifier needs to be removed.
What needs to happen really is that Objective becomes ObjectiveViewModel, and we convert what comes back to an Objective domain object (which may have more properties than we need for this screen). This can have a private set.
We're using RavenDB on the backend and so all the DB keys are strings that contain forward slashes e.g. users/1.
AFAIK there's no way to use the default ASP NET Web API route matching engine to match a parameter like this, without using a catch-all which means the key must be the last thing on the URL. I tried adding a regex constraint of users/d+ but it didn't seem to make a difference, the route wasn't matched.
What bits would I have to replace to do just enough custom route matching to allow this, preferably without crippling the rest of the route matching. For example, using url: "{*route}" and a custom constraint that did full regex matching would work but may work unexpectedly with other route configurations.
If your answer comes with some sample code, so much the better.
It seems that it is possible to do this by defining a custom route. In MVC4 (last stable released 4.0.30506.0), it is not possible to do by implementing IHttpRoute as per specification but by defining a custom MVC-level Route and adding it directly to the RouteTable. For details see 1, 2. The RegexRoute implementation below is based on the implementation here with mods from the answer here.
Define RegexRoute:
public class RegexRoute : Route
{
private readonly Regex urlRegex;
public RegexRoute(string urlPattern, string routeTemplate, object defaults, object constraints = null)
: base(routeTemplate, new RouteValueDictionary(defaults), new RouteValueDictionary(constraints), new RouteValueDictionary(), HttpControllerRouteHandler.Instance)
{
urlRegex = new Regex(urlPattern, RegexOptions.Compiled);
}
public override RouteData GetRouteData(HttpContextBase httpContext)
{
string requestUrl = httpContext.Request.AppRelativeCurrentExecutionFilePath.Substring(2) + httpContext.Request.PathInfo;
Match match = urlRegex.Match(requestUrl);
RouteData data = null;
if (match.Success)
{
data = new RouteData(this, RouteHandler);
// add defaults first
if (null != Defaults)
{
foreach (var def in Defaults)
{
data.Values[def.Key] = def.Value;
}
}
// iterate matching groups
for (int i = 1; i < match.Groups.Count; i++)
{
Group group = match.Groups[i];
if (group.Success)
{
string key = urlRegex.GroupNameFromNumber(i);
if (!String.IsNullOrEmpty(key) && !Char.IsNumber(key, 0)) // only consider named groups
{
data.Values[key] = group.Value;
}
}
}
}
return data;
}
}
Add this DelegatingHandler to avoid a NullReference due to some other bug:
public class RouteByPassingHandler : DelegatingHandler
{
protected override System.Threading.Tasks.Task<HttpResponseMessage> SendAsync(HttpRequestMessage request, System.Threading.CancellationToken cancellationToken)
{
HttpMessageInvoker invoker = new HttpMessageInvoker(new HttpControllerDispatcher(request.GetConfiguration()));
return invoker.SendAsync(request, cancellationToken);
}
}
Add handler and route directly to the RouteTable:
RouteTable.Routes.Add(new RegexRoute(#"^api/home/index/(?<id>\d+)$", "test", new { controller = "Home", action = "Index" }));
config.MessageHandlers.Add(new RouteByPassingHandler());
Et voila!
EDIT: This solution has problems when the API is self-hosted (instead of using a WebHost) and requires further work to make it work with both. If this is interesting to anyone, please leave a comment and I'll post my solution.
I have written an attribute before, but I I have not written a validation attribute before. I am seriously confused about how it all works together. I have read most of the tutorials online about how to go about accomplishing this. But I am left with a couple of questions to ponder.
Keep in mind that I am trying to write a requiredIf attribute that will only call a remote function if a certain Jquery variable is set... which incidentally is a variable that is pulled from view state... I guess I could make that part of my view model. But I digress
1) The C# code is slightly confusing. I know my attribute should extend the ValidationAttribute, IClientValidatable class and interface respectively. But I am a little confused about what each of the overidden methods should be doing? I am trying to write a requiredIf, how does overwriting these methods help me accomplish this goal?
2) If the variable is not there, I simply don't want the remote function to attempt to validate the field. I don't want any message to pop up on my form. Alot of the tutorials seem to revolve around that.
3) I am confused about what I need to do with the jquery to add this function to the view... What do I need to add to the JQuery to get this thing to work... It seems like a lot of extra coding when I could simply just type up a jquery function that did the same thing with just the same ore less coding... I know it also adds server side validation which is good. But still...
Here is what I have for my jquery side of this equation...
(function ($) {
$validator.unobtrusive.adapters.addSingleVal("requiredifattribute", "Dependent");
$validator.addMethod("requiredifattribute", function (value, element, params) {
if (!this.optional(element)) {
var otherProp = $('#' + params)
return (otherProp.val() != value);
}
return true;
})
}(jQuery));
Here is my Attribute (which is basically carbon copied out of one the required if tutorials... I know I need to customize it more, but once I get a better idea of what every piece is doing I will do that...
[AttributeUsage(AttributeTargets.Property)]
public class RequiredIfAttribute : ValidationAttribute, IClientValidatable {
private const string errorMessage = "The {0} is required.";
//public string
private RequiredAttribute innerAttribute = new RequiredAttribute();
public string DependentProperty { get; set; }
public object TargetValue { get; set; }
public RequiredIfAttribute(string dependentProperty, object targetValue){
this.DependentProperty = dependentProperty;
this.TargetValue = targetValue;
}
protected override ValidationResult IsValid(object value, ValidationContext validationContext) {
var field = validationContext.ObjectInstance.GetType().GetProperty(DependentProperty);
if (field != null) {
var dependentValue = field.GetValue(validationContext.ObjectInstance, null);
if ((dependentValue == null && TargetValue == null) || (dependentValue.Equals(TargetValue))) {
if (!innerAttribute.IsValid(value))
return new ValidationResult(ErrorMessage);
}
}
return ValidationResult.Success;
}
public IEnumerable<ModelClientValidationRule> GetClientValidationRules(ModelMetadata metadata, ControllerContext context) {
ModelClientValidationRule modelClientValidationRule = new ModelClientValidationRule {
ErrorMessage = FormatErrorMessage(metadata.DisplayName),
ValidationType = "requiredifattribute"
};
modelClientValidationRule.ValidationParameters.Add("dependent", DependentProperty);
yield return modelClientValidationRule;
}
}
UPDATE: What I have simply isn't working
Here is how a property in my model is anotated with the above attribute
[RequiredIf("isFlagSet", true)]
[Remote("ValidateHosFin", "EditEncounter", AdditionalFields = "hospitalFin, encflag", ErrorMessage = "Got Damn this is complex!")]
[MinLength(6)]
public string HostpitalFinNumber { get; set; }
The value in my view that I was trying to key this validation on is set up like so...
ViewData["ADDENCOREDITTEMP"] = encflag;
if (encflag == "AddEnc"){
isFlagSet = true;
}
I embed it into my page like so...
#Html.Hidden("isFlagSet", isFlagSet, new { id = "isFlagSet"})
I can't get my form to submit... The person who said he just tried this and got it to work, could you post the code?
Model:
public class X
{
[RequiredIf("y", "y", ErrorMessage = "y is not y")]
public string x { get; set; }
public string y { get; set; }
}
View:
#using(Html.BeginForm())
{
#Html.ValidationSummary()
#Html.TextBoxFor(m => m.x)
#Html.TextBoxFor(m => m.y)
<input type="submit"/>
}
I assume your validation fails on the server side? do you have isFlagSet property in your view model?
In this code from Microsoft's MVC Tutorial NerdDinners:
public class DinnerRepository {
private NerdDinnerDataContext db = new NerdDinnerDataContext();
//
// Query Methods
public IQueryable<Dinner> FindAllDinners() {
return db.Dinners;
}
public IQueryable<Dinner> FindUpcomingDinners() {
return from dinner in db.Dinners
where dinner.EventDate > DateTime.Now
orderby dinner.EventDate
select dinner;
}
public Dinner GetDinner(int id) {
return db.Dinners.SingleOrDefault(d => d.DinnerID == id);
}
//
// Insert/Delete Methods
public void Add(Dinner dinner) {
db.Dinners.InsertOnSubmit(dinner);
}
public void Delete(Dinner dinner) {
db.RSVPs.DeleteAllOnSubmit(dinner.RSVPs);
db.Dinners.DeleteOnSubmit(dinner);
}
//
// Persistence
public void Save() {
db.SubmitChanges();
}
}
What does:
public Dinner GetDinner(int id) {
return db.Dinners.SingleOrDefault(d => d.DinnerID == id);
}
the "d" mean? How does this code work? I know it it bringing back dinners where dinnerid matches id from the function parameter. I don't understand the "d goes to..." means. I know it is a lambda but I don't really get it. What is the "d" for? What does it do?
Could this have been written without the lambda here (how)?
You should probably read up on anonymous methods.
Basically the code you are referring to can be written as an anonymous method without lamba syntax like this:
public Dinner GetDinner(int id) {
return db.Dinners.SingleOrDefault(delegate (Dinner d) {
return d.DinnerID == id;
});
}
This is similar too...
from d in db.Dinners
where d.DinnerID == id
select d
The code basically loops around the dinners returning the first Dinner or the default if none is found.
This is a case where the naming conventions used in a sample aren't always appropriate in production. Using a "d" as a local variable is usually fround upon and choosing a variable name of "dinner" would probably be more appropriate, although in this case the scope of d is so small it is clear either way, as long as you know how lambda expressions work.
You need to understand lambda syntax and what it's used for.
Here's an article that does a decent job of explaining it.
However, to shortly answer your question in regards to the NerdDinner context, "d" in this context is just a parameter passed into the lamda expression that is a Dinner object.
This bit of code:
d => d.DinnerID == id
Can be thought of as defining a function of type Func<Dinner, bool>.
Whatever you pass it to, this function can be called, and passed Dinner, and it will give back a bool.
void Foo(Func<Dinner, bool> f)
{
bool result = f(new Dinner());
}
In your real example, the function SingleOrDefault will call the function you give it multiple times, passing it a different Dinner each time, and will return the Dinner for which the function returns true.
In fact, as you're using IQueryable, this is only conceptually what happens. The chances are, the code of the function is converted into SQL, and all the execution is done inside the database.