FluentValidation Validate Method throwing an excepion - c#

I'm having some issues the using FluentValidation library.
I have a very small Model
`
[FluentValidation.Attributes.Validator(typeof(PersonValidator))]
public class PersonModel
{
public string Name { get; set; }
public Nullable<short> Type { get; set; }
}
`
Have a validator class
public class PersonValidator : AbstractValidator<PersonModel>
{
public PersonValidator()
{
RuleFor(x => x.Name)
.Length(1, 5)
.WithLocalizedMessage(() => BaseValidationResource.LengthValidationMessage, 1, 5);
}
}
And I have a controller
public ActionResult Index()
{
var model = new PersonModel();
model.Name = "John Doe";
var validator = new PersonValidator();
var results = validator.Validate(model);
var error = GetModelErrors();
return View(model);
}
So far so good, the issue is that when the progam is executing and it gets to the line with ; var results = validator.Validate(model); it throws a SystemFormatException.
Instead of throwing an exception, shouldn't the validate method just return an object containing a boolean field which indicates if the model is valid and a list of errors.
PS : I know that this particular validation can also be done using DataAnnotations but i want to use Fluentvalidation because its more flexible.
Thanks in advance for your help.

As #RIc pointed out was a string formatting issue on my Resource file.
On my validor i had this line
RuleFor(x => x.Name)
.Length(1, 5)
.WithLocalizedMessage(() => BaseValidationResource.LengthValidationMessage, 1, 5);
Which pointed to the resource file and passed 2 parameters. But on my resource file i was expecting 3 parameters (property name, min value, max value).
However i was using the wrong annotation. Below are the before and after version of the the resource file.

Related

How can I add validations to print my own error message

I am creating a customer model for my API and I've set my 'Name' and 'Email' fields as required. But when we leave those fields empty while creating a new customer, we got a built-in error from EF core , instead I want to show my own error message...how can I do that.
I've tried to add validations by code in my post method but It doesn't work...can anyone help me with that?...Thanks in advance
If you want to do simple validation: Fluent Validation
public class UserRegisterValidator : AbstractValidator<User>
{
public UserRegisterValidator()
{
RuleFor(r => r.email).NotEmpty().WithMessage("Email address cannot be empty.");
.
.
.
}
}
When adding a new user
UserRegisterValidator v = new UserRegisterValidator();
ValidationResult result = v.Validate(usr);
if (!result.IsValid) return StatusCode(StatusCodes.Status400BadRequest, result.Errors); //Returns the specified error message
if (result.IsValid)
{
//If there is no error, the necessary actions are here
}
If you want to use server side validation you can specify the error message in validation tags. For example:
public class Movie
{
public int ID { get; set; }
[Required(ErrorMessage = "Name is required")]
public string Name{ get; set; }
}
Then you can also add some DetailsProvider to get better details when calling the api:
var builder = WebApplication.CreateBuilder(args);
builder.Services.AddControllers(options =>
{
options.ModelMetadataDetailsProviders.Add(new NewtonsoftJsonValidationMetadataProvider());
}).AddNewtonsoftJson();
Validation in front end has some other scenarios. Take a look at
https://learn.microsoft.com/en-us/aspnet/core/mvc/models/validation?view=aspnetcore-7.0

FluentValidation.AspNetCore is not working in Web API project

I am using nuget package FluentValidation.AspNetCore 11.0.1. I want to validate my model based on a condition for my endpoint. Added required code blocks as per the Fluent validation documentation, but still, my validation does not work.
I want to return a validation error when my model property contains letter "a" as in my validator class. Please check the below code blocks added to my project for validation.
Validator class
public class TestModelValidate : AbstractValidator<TestModel>
{
public TestModelValidate()
{
RuleFor(t => t.Summary.ToString()).NotEmpty().NotNull().When(x => x.Summary.Contains("a")).WithMessage("Cannot be null");
}
}
Startup class
Added below code block to ConfigureServices() method in Startup class.
services.AddControllers().AddFluentValidation(fvc => fvc.RegisterValidatorsFromAssemblyContaining<TestModelValidate>());
Model class
public class TestModel
{
public string Summary { get; set; }
}
Controller class
[HttpPost]
public IActionResult Test(TestModel model)
{
if (!ModelState.IsValid)
{
return new BadRequestObjectResult(ModelState);
}
return Ok("SUCCESS");
}
I am passing following JSON object to the endpoint using Postman.
{
"summary": "a"
}
Actual Result - SUCCESS
Expected Result - Validation Error
Appreciate, your help!
If you want to use ModelState.IsValid we can try to set ApiBehaviorOptions.SuppressModelStateInvalidFilter be true.
builder.Services.Configure<ApiBehaviorOptions>(options =>
{
options.SuppressModelStateInvalidFilter = true;
});
I think you can try to use Must with the condition !x.Contains("a") to judge your logic.
public class TestModelValidate : AbstractValidator<TestModel>
{
public TestModelValidate()
{
RuleFor(t => t.Summary).NotEmpty().NotNull().WithMessage("Cannot be null")
.Must(x => !x.Contains("a")).WithMessage("Can't contain a");
}
}

FluentAssertions Equivalency Comparison Behavior and IMemberInfo

I am using FluentAssertions (v6.2.0) to test API's that return table-like data. I want to change comparison behavior for one of the field, and tried to use method described in documentation.
orderDto.Should().BeEquivalentTo(order, options => options
.Using<DateTime>(ctx => ctx.Subject.Should().BeCloseTo(ctx.Expectation, 1.Seconds()))
.When(info => info.Name == "Date"));
The issue is that IMemberInfo class that When extension method is expecting doesn't have Name property, it has property called Path. Was Name replaced by Path and this is a typo in documentation, or do I need to import another namespace to use Name property?
From a quick look at the FluentAssertions source code, I'm seeing that the info argument is of type IObjectInfo and it has a Path property. A quick test with this code shows the Path property working as you would expect:
void Main()
{
var orderDto = new OrderDto { Date = DateTime.Now };
var order = new Order { Date = DateTime.Now };
//var order = new Order { Date = DateTime.Now.Subtract(TimeSpan.FromSeconds(2)) };
orderDto.Should().BeEquivalentTo(order, options => options
.Using<DateTime>(ctx => ctx.Subject.Should().BeCloseTo(ctx.Expectation, 1.Seconds()))
.When(info => info.Path == "Date"));
}
class OrderDto {
public DateTime Date { get; set; }
}
class Order
{
public DateTime Date { get; set; }
}
In fact, the FluentAssertions test for that code also uses path. See https://github.com/fluentassertions/fluentassertions/blob/master/Tests/FluentAssertions.Specs/Equivalency/ExtensibilityRelatedEquivalencySpecs.cs#L394
There is also an IMethodInfo interface with both Name and Path properties. However, that is used by the Include* and Exclude* methods.
So it appears to be a documentation bug.

Fluent Validation output current record information

Ok, I am using Fluent Validation for one of my classes what I want to know is. How does one determine what record is at fault i.e say for example the following
Is being classed as a number that the customer can refer to how would I Change my string below using Fluent Validation to output the current record it's working on Document No to the customer.
public string DocumentNo { get; set; }
Is it just as simple as appending it to the string?
Code:
public class SupplierTransactionsValidation : AbstractValidator<SageStandardImportInvoces>
{
public SupplierTransactionsValidation()
{
RuleFor(x => x.AnalysisCode1) // code repeated
.NotEqual("None").WithMessage("Please enter a value for AnalysisCode1")
.Length(0, 3);
RuleFor(x => x.AnalysisCode2) // code repeated
.NotEqual("None").WithMessage("Please enter a value for AnalysisCode2")
.Length(0, 3);
RuleFor(x => x.AnalysisCode3) // code repeated
.NotEqual("None").WithMessage("Please enter a value for AnalysisCode3")
.Length(0, 3);
}
}
If I understand your issue correctly, you could create a private method that gets the name of the property to be validated by casting the body of the expression to a MemberExpression:
public class SupplierTransactionsValidation : AbstractValidator<SageStandardImportInvoces>
{
public SupplierTransactionsValidation()
{
BuildRule(x => x.AnalysisCode1);
BuildRule(x => x.AnalysisCode2);
BuildRule(x => x.AnalysisCode3);
}
private IRuleBuilderOptions<SageStandardImportInvoces, string>
BuildRule(System.Linq.Expressions.Expression<Func<SageStandardImportInvoces, string>> expression)
{
return RuleFor(expression)
.NotEqual("None")
.WithMessage($"Please enter a value for {(expression.Body as System.Linq.Expressions.MemberExpression)?.Member.Name}")
.Length(0, 3);
}
}
This way you don't have to repeat your logic.

How can I reuse a DropDownList in several views with .NET MVC

Several views from my project have the same dropdownlist...
So, in the ViewModel from that view I have :
public IEnumerable<SelectListItem> FooDdl { get; set; }
And in the controller I have :
var MyVM = new MyVM() {
FooDdl = fooRepository.GetAll().ToSelectList(x => x.Id, x => x.Name)
}
So far so good... But I´m doing the same code in every view/controller that have that ddl...
Is that the best way to do that?
Thanks
I'd say that's fine to be honest, as it's only a repeat of a few lines of code. If it's really bothering you though, you could have all your controllers inherit from a BaseController (if they don't already) and store a method in there to get them all, something like:
public IEnumerable<SelectListItem> GetFoos()
{
return fooRepository.GetAll().ToSelectList(x => x.Id, x => x.Name);
}
Then in your controllers you could do:
var MyVM = new MyVM() {
FooDdl = GetFoos()
}
If your DropDownList is exactly the same the approach I would use is:
1) In your Base Controller or in a Helper class, you can create a method that returns a SelectList. That method should receive a nullabe int to get the select list with a value pre selected.
2) It is wise to cache the information you list in the DDL, to not query the database too often.
So, for (1):
public SelectList GetMyDDLData(int? selectedValue){
var data = fooRepository.GetAll().Select(x => new { Value = x.Id, Text = x.Name });
return new SelectList(data, "Id","Name", selectedValue);
}
In the view model:
var myVM = new MyVM();
myVM.DDLData = this.GetMyDDLData(null) // if it is in your BaseController.
myVM.DDLData = YourHelperClass.GetMyDDLData(null) // if it is in a helper static class
In your views:
#Html.DropDownListFor(x => x.FooProp, Model.DDLData, "Select one...")
For number (2):
private IEnumerable<YourModel> GetMyData()
{
var dataItems = HttpContext.Cache["someKey"] as IEnumerable<YourModel>;
if (dataItems == null)
{
// nothing in the cache => we perform some expensive query to fetch the result
dataItems = fooRepository.GetAll().Select(x => new YourModel(){ Value = x.Id, Text = x.Name };
// and we cache it so that the next time we don't need to perform the query
HttpContext.Cache["someKey"] = dataItems ;
}
return dataItems;
}
The "someKey" could be something specific and static is this data is the same to all users, or you can do "someKey" + User.Id if the data is specific to one user.
If your repository is an abstractin layer (not directly EntityFramework) you can place this code there.
We also use a static class :
public static class SelectLists
{
public static IList<SelectListItem> CompanyClasses(int? selected)
{
var session = DependencyResolver.Current.GetService<ISession>();
var list = new List<SelectListItem>
{
new SelectListItem
{
Selected = !selected.HasValue,
Text = String.Empty
}
};
list.AddRange(session.All<CompanyClass>()
.ToList()
.OrderBy(x => x.GetNameForCurrentCulture())
.Select(x => new SelectListItem
{
Selected = x.Id == (selected.HasValue ? selected.Value : -1),
Text = x.GetNameForCurrentCulture(),
Value = x.Id.ToString()
})
.ToList());
return list;
}
}
In the view we have nothing special :
#Html.DropDownListFor(x => x, SelectLists.CompanyClasses(Model))
And sometime we also create an EditorTemplate so it's faster to reuse like this
Model :
[Required, UIHint("CompanyClassPicker")]
public int? ClassId { get; set; }
EditorTemplate :
#model int?
#if (ViewBag.ReadOnly != null && ViewBag.ReadOnly)
{
var item = SelectLists.CompanyClasses(Model).FirstOrDefault(x => x.Selected);
if (item != null)
{
<span>#item.Text</span>
}
}
else
{
#Html.DropDownListFor(x => x, SelectLists.CompanyClasses(Model))
}
Create object with getter for your dropdown values:
public static class DropDowns
{
public static List<SelectListItem> Items {
get
{
//Return values
}
}
}
Create Razor partial:
#Html.DropDownListFor(m => "ChoosenItem", DropDowns.Items, "")
Call partial:
#Html.RenderPartial("DropDownItems")
And finally receive ChoosenItem value in controller. Simply.
I use an IModelEnricher combined with Automapper and attributes that define relationships between a type of list and select list provider. I return an entity etc using a specific ActionResult that then automaps my entity to a ViewModel and enriches with data required for select lists (and any additional data required). Also keeping the select list data as part of your ViewModel keeps your controller, model, and view responsibilities clear.
Defining a ViewModel ernicher means that anywhere that ViewModel is used it can use the same enricher to get its properties. So you can return the ViewModel in multiple places and it will just get populated with the correct data.
In my case this looks something like this in the controller:
public virtual ActionResult Edit(int id)
{
return AutoMappedEnrichedView<PersonEditModel>(_personRepository.Find(id));
}
[HttpPost]
public virtual ActionResult Edit(PersonEditModel person)
{
if (ModelState.IsValid){
//This is simplified (probably don't use Automapper to go VM-->Entity)
var insertPerson = Mapper.Map<PersonEditModel , Person>(person);
_personRepository.InsertOrUpdate(insertPerson);
_requirementRepository.Save();
return RedirectToAction(Actions.Index());
}
return EnrichedView(person);
}
This sort of ViewModel:
public class PersonEditModel
{
public string FirstName { get; set; }
public string LastName { get; set; }
public int Age { get; set; }
public int FavouriteTeam { get; set; }
public IEnumerable<SelectListItem> Teams {get;set;}
}
With this sort of Enricher:
public class PersonEditModelEnricher :
IModelEnricher<PersonEditModel>
{
private readonly ISelectListService _selectListService;
public PersonEditModelEnricher(ISelectListService selectListService)
{
_selectListService = selectListService;
}
public PersonEditModelEnrich(PersonEditModel model)
{
model.Teams = new SelectList(_selectListService.AllTeams(), "Value", "Text")
return model;
}
}
One other option is to decorate the ViewModel with attributes that define how the data is located to populate the select list. Like:
public class PersonEditModel
{
public string FirstName { get; set; }
public string LastName { get; set; }
public int Age { get; set; }
public int FavouriteTeam { get; set; }
[LoadSelectListData("Teams")]
public IEnumerable<SelectListItem> Teams {get;set;}
}
Now you can decorate an appropriate method in your select service with an attribute like:
[ProvideSelectData("Teams")]
public IEnumerable Teams()
{
return _teamRepository.All.ToSelectList(a => a.Name, a => a.TeamId);
}
Then for simple models with no complex enrichment just the generic enrichment process can handle it. If you want to do anything more complex you can define an enricher and it will be used if it exists.
Other options could be a convention over configuration approach where the Enricher looks at property name and type e.g. IEnumerable<SelectListItem> PossibleFirstDivisionTeams {get;set;} then matches this if it exists with a select list provider name in a class that say implements a marker interface ISelectListProvider. We went the attribute based one and just created Enums representing the various lists E.g. SelectList.AllFirstDivisionTeams. Could also try interfaces on ViewModel that just have a property collection for a selectlist. I don't really like interfaces on my ViewModels so we never did this
It all really depends on the scale of your application and how frequently same type of select list data is required across multiple models. Any specific questions or points you need clarified let me know
See this question. Also this blog post and this. Also this question on Automapper forum
The first question is if the options-list belongs to the ViewModel. A year or two ago I did the same, but what I see recently more and more as a "best practice" is that people add the list to the ViewBag/ViewData not to the ViewModel. That's an option and I tend to do the same for a one-shot drop-down list, but it doesn't answer the code-reuse question you are facing. For that I see two different approaches (and two more that I rule out).
Shared editor template. Create an editor template for the type that's represented by the dropdown. In this case - because we don't have the list of possible options in the ViewModel or the ViewBag - the template has to reach out for the options to the server. That's possible by adding an action method (that returns json) to a controller class. Either to a shared "LookupsController" (possibly an ApiController) or to the controller that the list-items' type belongs to.
Partial view. The drop down values belong to some type. The Controller of that type could have an action method that returns a partial view.
The benefit of the first one is that a nice #Html.EditorFor call will do the job. But I don't like the ajax dependency. Partly for that reason I would prefer the partial view.
And there is a third one: child action, but I don't see that a good pattern here. You can google what's the difference between child actions and partial views, for this case child action is the wrong choice. I also wouldn't recommend helper methods. I believe they are not designed for this use case either.
You could put that fetch in the default (null) constructor of MyVM if you don't need to vary its content.
Or you could use a PartialView that you render into the views that need t.
I like to use static classes often in a helper class that I can call from any view.
#Html.DropDownListFor(x => x.Field, PathToController.GetDropDown())
and then in your controller have a method built like this
public static List<SelectListItem> GetDropDown()
{
List<SelectListItem> ls = new List<SelectListItem>();
lm = (call database);
foreach (var temp in lm)
{
ls.Add(new SelectListItem() { Text = temp.name, Value = temp.id });
}
return ls;
}
Hopefully it helps.
If you really don't want to duplicate the code, place the code from the controllers into a helper class, and render the dropdown within a shared view (like _Layout.cshtml) that you'd then have to implement into your views by RenderPartial.
Create a partial view, _MyDropdownView.cstml, which uses the helper class you threw the code from the controllers in with something like the following:
#using MyNamespace.MyHelperClass
<div id="myDropdown">#Html.DropDownListFor(model => model.Prop, MyVM as SelectList, "--Select a Property--")</div>
Then, within your views:
#Html.RenderPartial("_MyDropdownView")
Extension methods to the rescue
public interface ISelectFoo {
IEnumerable<SelectListItem> FooDdl { get; set; }
}
public class FooModel:ISelectFoo { /* implementation */ }
public static void PopulateFoo(this ISelectFoo data, FooRepository repo)
{
data.FooDdl = repo.GetAll().ToSelectList(x => x.Id, x => x.Name);
}
//controller
var model=new ViewModel();
model.PopulateFoo(repo);
//a wild idea
public static T CreateModel<T>(this FooRepository repo) where T:ISelectFoo,new()
{
var model=new T();
model.FooDdl=repo.GetAll().ToSelectList(x => x.Id, x => x.Name);
return model;
}
//controller
var model=fooRepository.Create<MyFooModel>();
What about a Prepare method in a BaseController?
public class BaseController : Controller
{
/// <summary>
/// Prepares a new MyVM by filling the common properties.
/// </summary>
/// <returns>A MyVM.</returns>
protected MyVM PrepareViewModel()
{
return new MyVM()
{
FooDll = GetFooSelectList();
}
}
/// <summary>
/// Prepares the specified MyVM by filling the common properties.
/// </summary>
/// <param name="myVm">The MyVM.</param>
/// <returns>A MyVM.</returns>
protected MyVM PrepareViewModel(MyVM myVm)
{
myVm.FooDll = GetFooSelectList();
return myVm;
}
/// <summary>
/// Fetches the foos from the database and creates a SelectList.
/// </summary>
/// <returns>A collection of SelectListItems.</returns>
private IEnumerable<SelectListItem> GetFooSelectList()
{
return fooRepository.GetAll().ToSelectList(foo => foo.Id, foo => x.Name);
}
}
You can use this methods in the controller:
public class HomeController : BaseController
{
public ActionResult ActionX()
{
// Creates a new MyVM.
MyVM myVm = PrepareViewModel();
}
public ActionResult ActionY()
{
// Update an existing MyVM object.
var myVm = new MyVM
{
Property1 = "Value 1",
Property2 = DateTime.Now
};
PrepareViewModel(myVm);
}
}
Have an interface with all your properties that need to be automatically populated:
public interface ISelectFields
{
public IEnumerable<SelectListItem> FooDdl { get; set; }
}
Now all your view models that want to have those properties, implement that interface:
public class MyVM : ISelectFields
{
public IEnumerable<SelectListItem> FooDdl { get; set; }
}
Have a BaseController, override OnResultExecuting, find the ViewModel that is passed in and inject the properties to the interface:
public class BaseController : Controller
{
protected override void OnResultExecuting(ResultExecutingContext filterContext)
{
var viewResult = filterContext.Result as ViewResult;
if (viewResult != null)
{
var viewModel = viewResult.Model as ISelectFields;
if (viewModel != null)
{
viewModel.FooDdl = fooRepository.GetAll().ToSelectList(x => x.Id, x => x.Name)
}
}
base.OnResultExecuting(filterContext);
}
}
Now your controllers are very simple, everything is strongly typed, you are sticking with the DRY principle and you can just forget about populating that property, it will always be available in your views as long as your controllers inherit from the BaseController and your ViewModels implement the interface.
public class HomeController : BaseController
{
public ActionResult Index()
{
MyVM vm = new MyVM();
return View(vm); //you will have FooDdl available in your views
}
}
Why not use the advantages of RenderAction:
#(Html.RenderAction("ControllerForCommonElements", "CommonDdl"))
Create a controller, and an action that returns the Ddl and and just reference it in the views.
See some tips here on how you could use it
This way you can also cache this result. Actually the guys building StackOverflow talked about the pros of using this combined with different caching rules for different elements (i.e. if the ddl does not need to be 100% up to date you could cache it for a minute or so) in a podcast a while ago.

Categories