I have a REST interface endpoint like
POST /items/12345/actions
I utilize a generic actions sub collection to be apply to apply changes to 12345 which are not easily mapped to the content or direct other sub collections of it.
My question is the following: Since there could be multiple different action types I identify the action by a JSON property of the content of an uploaded document.
How do I select a action by a part of the JSON body of the request. Is there something possible like...
[Route("api/v1/items")
public class ItemsController : Controller
{
[HttpPost("{id}/actions")]
[CheckJsonBody("type", "ActionA")]
public ActionResult DoActionA(int id, ActionA a)
{
// do something
}
[HttpPost("{id}/actions")]
[CheckJsonBody("type", "ActionB")]
public ActionResult DoActionB(int id, ActionB b)
{
// do something
}
}
The request would look like ...
{
"type": "ActionA",
"abc": "xyz"
}
I have digged myself up into the code till Microsoft.AspNetCore.Mvc.ActionConstraints.ActionMethodSelectorAttribute (GitHub).
However starting from there, I am a bit lost to reach a high-performance solution. Do I need to decode the body or is that something which is already done at that time the constraint is evaluated?
ps: And yes, I know I could handle them in one action and do a switch on the "type" property.
An ASP.NET team member was so friendly to direct me to an answer: In the ActionMethodSelectorAttribute you can read the body into a memory stream, read till the property for the selection filter. Then you seek the memory stream to zero and replace it in the request (for later model binding). You can cache the criteria value in HttpContext.Items to speed it up if you use the same property for multiple actions.
Related
Is it possible to setup Web Api 2 route based on a parameter's value in the query string.
I have the following requirement:
/api/controller?data=param.data1
should go to controller's action 1
/api/controller?data=param.data2
should go to controller's action 2
any other value of data must go to action 3.
I know there's an option to set a constraint with a regex, but the examples I've found are for generic scenarios and not as specific as mine.
This is what I've tried
config.Routes.MapHttpRoute(
name: "test",
routeTemplate: "api/Hub/{data2}",
defaults: new { action = "Test" },
constraints: new { data2 = #"^(param\.data2)$" }
);
Is there a way to do it? Maybe there's a better way?
Important to note, I cannot change the URI of the service. It must have ?data=[value]
This is a fallback for a legacy system :(
You can use Attribute Routing, new in Web API 2.
Let's say you have the following actions, where the data param is, let's say, a string:
public Stuff GetStuffForData1(string data) { ... }
public Stuff GetStuffForData2(string data) { ... }
public Stuff GetStuffForData(string data) { ... }
Since you mentioned regex, you can specify route constraints for each of the above actions using a regex like the one you mentioned in your question1, for example:
[Route("controller/{data:regex(#"^(param\.data1)$")]
public Stuff GetStuffForData1(string data) { ... }
[Route("controller/{data:regex(#"^(param\.data2)$")]
public Stuff GetStuffForData2(string data) { ... }
// No need for a route constraint for other data params.
public Stuff GetStuffForData(string data) { ... }
The general syntax is {parameterName:constraint(params)} (params is optional and is not used for all constraints). In the above example, the first route will only be selected if the data segment of the URI matches the data1 regex. Similarly, the second route will be selected if the data segment of the URI matches the data2 regex. Otherwise, the last route will be chosen.
In general, the total ordering is determined as follows:
Compare the RouteOrder property of the route attribute. Lower values are evaluated first. The default order value is zero.
Look at each URI segment in the route template. For each segment, order as follows:
Literal segments.
Route parameters with constraints.
Route parameters without constraints.
Wildcard parameter segments with constraints.
Wildcard parameter segments without constraints.
In the case of a tie, routes are ordered by a case-insensitive ordinal string comparison (OrdinalIgnoreCase) of the route template.
You can even create your own custom route constraints by implementing the IHttpRouteConstraint interface and registering it in the Register method of your WebApiConfig class, assuming you're hosting on IIS, or in the Configuration method of your Startup class if self-hosting using OWIN.
Note I haven't personally tried any of the above, but it should all work; at the very least it should give you some ideas. For more details, including very nice examples, you should start by taking a look at the following article (which I shamelessly used extensively in my answer):
http://www.asp.net/web-api/overview/web-api-routing-and-actions/attribute-routing-in-web-api-2#constraints
1 I'm really not an expert on writing regexes, so unfortunately I can't advise you on the specific ones you'll need.
I'm exploring using FluentValidation as it seems to be an elegant API for validation of my ViewModels upon model binding. I'm looking for opinions on how to properly centralize validation using this library as well as from my business (service) layer and raise it up to the view without having 2 different approaches to adding modelstate errors.
I'm open to using an entirely different API but essentially looking to solve this branching validation strategy.
[Side Note: One thing I tried was to move my business method into my FluentValidation's custom RsvpViewModelValidator class and using the .Must method but it seemed wrong to hide that call in there because if I needed to actually use my Customer object they I would have to re-query it again since its out of scope]
Sample Code:
[HttpPost]
public ActionResult AcceptInvitation(RsvpViewModel model)
{
//FluentValidation has happened on my RsvpViewModel already to check that
//RsvpCode is not null or whitespace
if(ModelState.IsValid)
{
//now I want to see if that code matches a customer in my database.
//returns null if not, Customer object if existing
customer = _customerService.GetByRsvpCode(model.RsvpCode);
if(customer == null)
{
//is there a better approach to this? I don't like that I'm
//splitting up the validation but struggling up to come up with a
//better way.
ModelState.AddModelError("RsvpCode",
string.Format("No customer was found for rsvp code {0}",
model.RsvpCode);
return View(model);
}
return this.RedirectToAction(c => c.CustomerDetail());
}
//FluentValidation failed so should just display message about RsvpCode
//being required
return View(model);
}
[HttpGet]
public ActionResult CustomerDetail()
{
//do work. implementation not important for this question.
}
To give some closure to the question (and make it acceptable) as well as summarize the comments:
Business/process logic and validation logic are two entities. Unless the validation ties in to the database (e.g. check for unique entries) there's no reason to group validation into one location. Some are responsible in the model making sure there's nothing invalid about the information, and some handle how the validated values are used within the system. Think of it in terms of property getters/setters vs the logic used in the methods with those properties.
That being said, separating out the processes (checks, error handling, etc.--anything not relating to UI) can be done in a service layer which also tends to keep the application DRY. Then the action(s) is/are only responsible for calling and presenting and not performing the actual unit of work. (also, if various actions in your application use similar logic, the checks are all in one location instead of throw together between actions. (did I remember to check that there's an entry in the customer table?))
Also, by breaking it down in to layers, you're keeping concerns modular and testable. (Accepting an RSVP isn't dependent on an action in the UI, but now it's a method in the service, which could be called by this UI or maybe a mobile application as well).
As far as bubbling errors up, I usually have a base exception that transverses each layer then I can extend it depending on purpose. You could just as easily use Enums, Booleans, out parameters, or simply a Boolean (the Rsvp either was or wasn't accepted). It just depends on how finite a response the user needs to correct the problem, or maybe change the work-flow so the error isn't a problem or something that the user need correct.
You can have the whole validation logic in fluent validation:
public class RsvpViewValidator : AbstractValidator<RsvpViewModel>
{
private readonly ICustomerService _customerService = new CustomerService();
public RsvpViewValidator()
{
RuleFor(x => x.RsvpCode)
.NotEmpty()
.Must(BeAssociatedWithCustomer)
.WithMessage("No customer was found for rsvp code {0}", x => x.RsvpCode)
}
private bool BeAssociatedWithCustomer(string rsvpCode)
{
var customer = _customerService.GetByRsvpCode(rsvpCode);
return (customer == null) ? false : true;
}
}
I am designing a REST API.
I have a single resource that I want to be able to change the status of for different conditions e.g. the URI is:
Applications/{application_id}/
The possible status changes are to set the application to:
Cancelled
SignedOff
Hold
Each status change will require different information e.g. a reason for cancelled, a date for signedoff.
What would be a good looking URI to handle this? I had thought of
POST: Applications/{application_id}/Cancel
POST: Applications/{application_id}/SignOff
POST: Applications/{application_id}/Hold
but it doesnt seem right to me.
EDIT:
I should have mentioned that I was already planning
POST: Applications/{application_id}
to update an existing application with a full set of application data.
I would stick with one url for all statuses and have your Status object encapsulate all the different properties. These keeps your url from having words that look like actions and to be more restful.
POST: Applications/{application_id}/status
public class Status
{
public string StatusType {get;set;}
public string CancelReason {get;set;}
public string SignOffDate {get;set;}
...
}
POST: Applications/{application_id}?cancel=true
POST is used only for CREATE. I think put will be better option.
I have a really weird scenario where I try to store domain events (I'm trying to learn CQRS and RavenDB at the same time). The basic structure of the documents I try to store are:
public interface IDomainEvent { ... }
public abstract class BaseDomainEvent : IDomainEvent { ... }
public class DomainEventA : BaseDomainEvent { ... }
public class DomainEventB : BaseDomainEvent { ... }
Given that I want to store DomainEventA and DomainEventB in the same collection in RavenDB and I have managed to do so. But the problem is that in the collection I am missing the properties of DomainEventB, and not all properties are set even though I have checked that the properties are set before I commit the transaction where I store the objects. The following gist shows a working example of what I want to do: https://gist.github.com/2830093, and the test code that fails me is found in this test: https://github.com/mastoj/TJ.CQRS/blob/ravenfail/TJ.CQRS.RavenEvent.Tests/RavenEventStoreTests.cs that is using this RavenDB code: https://github.com/mastoj/TJ.CQRS/blob/ravenfail/TJ.CQRS.RavenEvent/RavenEventStore.cs.
I really can't get my head around this one.
EDIT 1: I can add that in the failing scenario the metadata of the stored object says it is one type but the properties for that type is not stored.
I planned to delete or vote for close but I think more than me might experience this problem at some point. I found the solution in my case and it was that the objects I added to RavenDB had a faulty equals method so RavenDB thought that all my objects were the same one. When I added one more property to check in the equals method everything start working as expected.
I'm trying out the API controller in MVC 4 and have a question about best practice when using put, updating an object.
I'm using RavenDB as data storage and I have created a custom MediaTypeFormatter that has a converter of type JsonDynamicConverter that can serialize and deserialize json to a dynamic object. With that said my API controllers Put method looks like this
public void Put(string id, dynamic model) {}
this dynamic object looks sorta like this:
pageModel = {
"id": "pages-2",
"metadata": {
"changed": "2012-02-28T17:16:27.323Z"
},
"parent": {
"id": "pages-1",
"slug": null
},
"children": []
}
so, so far so good but now I need to update my entity with id pages-2.
The UpdateModel does not exist in the ApiController so my question is what is the best/preferred way of doing this?
Simply call RavenDB to load the object with the appropriate ID, make the changes to its contents and persist it again.
No need to worry about any UpdateModel calls. It doesn't apply here.
Be aware of one potential issue since you are including the id in the model. If I sent a PUT command to http://server/controller/pages-3 with that body what would happen? You should probably send me a bad request response or something similar depending on how you want your API to work.
Is there any reason not to be explicit in your parameters? I would define an UpdateModel and take that as parameter instead of the dynamic. Then it also would be able to model validation.
ASP.NET WebApi includes handling of converting from both JSON and XML as input to your controller methods. I'm guessing your combination of custom mediatypeformatter and use of dynamic would be unneccesary in this case, if there is not something I'm missing.