I did some research about this problem, but NOTHING comes close to MY difficulty.
Here's my SQL :
CREATE TABLE [dbo].[Departement] (
[IdDepartement] INT IDENTITY (1, 1) NOT NULL,
[Description] NVARCHAR (255) NOT NULL,
[IsActif] BIT NULL,
PRIMARY KEY CLUSTERED ([IdDepartement] ASC)
);
As you can see, IsActif IS null.
Here's part of my Departement.cs :
public Nullable<bool> IsActif { get; set; }
Once again, everything is normally nullable.
Here's part of my Metadata (for not loosing the annotations when the database is updated) :
[UIHint("YesNo")]
public Nullable<bool> IsActif { get; set; }
Still, a nullable force of the nature, without [Required]! (If someone wonders what is a UIHint: it is connected to a DisplayTemplate and it just presents nicely my bit value).
Just for the curious ones, because I feel my question will not be complete if I do not put everything in here. Here's my UiHint:
#model Nullable<bool>
#if (Model.HasValue)
{
if (Model.Value)
{ <text>Oui</text> }
else
{ <text>Non</text> }
}
else
{ <text>Indéfini</text> }
It clearly states if Model Has Value ON Nullable!!!
But this is what I get when a try to Create a new department:
which means 'the field IsActif is required.'
Someone has an idea what is going on? All my other nullable booleans work perfectly. Is there's a place I forgot to look? What's the next step?
If someone have an idea, please let me know. I did pet my computer to make this work, but I am about to eat it now.
Thanks in advance.
Edit
Mea Culpa: Oooooooooh, I so want to die. I just realised I was using a ViewModel and I should check there. And there it was, the culprit looking at me with a devilish snigger. I am sorry for this newbie mistake and for my punishment, I will close this question and mark it as answered and leave it here for everybody to laugh at me and just in case another newbie have the same problem. I still do not understand why my others nullable boolean worked, while they were in the same situation, though...
Just go see my Edit. I am too shameful to repeat it here. Thanks anyway to those who took their time to add comments, specially Jasen.
In big:
EmployeViewModel before:
public bool IsActif { get; set; }
EmployeViewModel after:
public Nullable<bool> IsActif { get; set; }
Now I just have to hide myself under a carpet.
Related
I'm coding an application using CQRS. I have my commands and my commandHandler. I do tests my commands handlers. But...should you recommend testing commands? What's the best strategy? Should I test all validation possibilities? How do you guys do?
using System;
using Flunt.Notifications;
using Flunt.Validations;
using Todo.Domain.Commands.Contracts;
namespace Todo.Domain.Commands
{
public class CreateTodoCommand : Notifiable, ICommand
{
public CreateTodoCommand() { }
public CreateTodoCommand(string title, string user, DateTime date)
{
Title = title;
User = user;
Date = date;
}
public string Title { get; set; }
public string User { get; set; }
public DateTime Date { get; set; }
public void Validate()
{
AddNotifications(
new Contract()
.Requires()
.HasMinLen(Title, 3, "Title", "Please, describe better your task!")
.HasMinLen(User, 6, "User", "Invalid User!")
);
}
}
}
should you recommend testing commands?
How often do you change commands?
How likely are you to introduce a mistake when changing a command?
How expensive is it for this mistake to escape into a release?
How expensive is it to test?
Commands are out near the input boundary, and one of the common problems that we have near the input boundary is parsing some general purpose data structure to ensure that it satisfies our bespoke constraints.
Does parsing need to be tested? Well, it is logic (more than one possible outcome), so it is certainly a candidate to consider.
Chances are: you don't have particularly specific answers to the questions above. In that case, I would err on the side of introducing tests: because the tests give you a way to collect data, which over time will improve your answers.
Is there a simple way to specify a list of possible values for the parameter orderBy? Not one by one please, otherwise I would not be making the question. I want to specify that orderby makes sense only if it is chosen from a predetermined list. Suppose the list is very large...still not random. This cannot be that hard...no single example of such a simple task.
[Test, AutoData]
public override void IndexReturnsView(int? pageIndex, int? pageSize, string orderBy, bool? desc)
{
.....
}
EDIT:
All I want is to read the possible values from a list as I would do with the ValueSource attribute. However, it seems not to work with AutoFixture. If I specified e.g. [ValueSource("GetOrderByColumnNames")] my test does not work anymore. I have no idea of what I am doing wrong. Unfortunately AutoFixture lacks useful documentation and the examples are very basic. Is there a working example of this scenario that I can use to guide myself here?
This has to be a very common situation, however I have been looking for days with no luck :(.
Appreciated!
If I understand the question correctly, the problem is that the orderBy value should be randomly selected from a list of predefined values, but that list might be too large to use with [InlineAutoData].
The easiest way to do this that I can think of is to introduce a helper type. This might actually be a valuable addition to the application code itself, as it makes the role of various values more explicit, but if not, you can always add the wrapper type to the test code base.
Something like this is the minimum you'll need:
public class OrderCriterion
{
public OrderCriterion(string value)
{
Value = value;
}
public string Value { get; }
}
If we also imagine that this class exposes a list of ValidValues, you can implement an AutoFixture Customization using the ElementsBuilder class:
public class OrderCriterionCustomization : ICustomization
{
public void Customize(IFixture fixture)
{
fixture.Customizations.Add(
new ElementsBuilder<OrderCriterion>(OrderCriterion.ValidValues));
}
}
Then you create a data source attribute for your test code base:
public class TestConventionsAttribute : AutoDataAttribute
{
public TestConventionsAttribute() : base(
() => new Fixture().Customize(new OrderCriterionCustomization()))
{
}
}
This enables you to write a test like this, which passes:
[Theory, TestConventions]
public void IndexReturnsView(
int? pageIndex,
int? pageSize,
OrderCriterion orderBy,
bool? desc)
{
Assert.Contains(orderBy.Value, OrderCriterion.ValidValues.Select(x => x.Value));
}
Notice that instead of declaring the orderBy parameter as a string, you declare it as an OrderCriterion, which means that AutoFixture will be detect its presence, and the Customization then kicks in.
See also https://stackoverflow.com/a/48903199/126014
I have a MVC4 application with unobstrusive javascript client-side validation. This is working fine except for decimal validation. I am encoutering some strange and annoying behaviour.
The decimal validation only allows numbers with three numbers after the decimal separator. However i want this to work with two numbers after the separator (or just any number is fine as it should do by default if i'm correct).
so 1.222 is valid and 1.22 and isn't while i wan't it to be valid. (1.2222 is neither valid).
in my view i have a normal:
#Html.TextBoxFor(x => x.BasePrice)
my model:
[Required]
public decimal BasePrice { get; set; }
and my controller:
ProductVM model = new ProductVM()
{
BasePrice = product.BasePrice
};
product is an Entity Framework 4 entityobject
I've also tried putting;
[DisplayFormat(DataFormatString = "{0:n2}", ApplyFormatInEditMode = true)]
in my model. But this doesn't work either. It has no effect.
I've never seen any behaviour like this when working with MVC. I have no idea where to look. Any clues?
I still don't know what the problem is, but I fixed it with a little bit dirty workaround:
$.validator.methods.number = function (value, element) {
if(!isNaN(parseFloat(strValue)))
{
return true;
}
return false;
}
by overriding the client side decimal validation. just added this in a script block in my _layout.cshtml.
server side validation will ensure the security is still ok by complete validation.
Can someone help me with my situation. I have the following code.
class Employee {
Integer empId
String firstName
String lastName
static constraints = {
empId()
firstName()
lastName()
}
static mapping = {
id generator:'assigned', name:'empId'
version false
}
}
The code allows me to save employee through 'create' but gives the following error message
"Employee not found with id null" and also in the list, all the employees are listed but clicking on any Emp Id gives the same error. Please help. This is driving me nuts.
Rocky
Thanks for your reply. Like I mentioned I am able to save the empId in the database but get that message nonetheless and see the list with assigned ids (empId). The link points to employee/show with no number at end. However employee/show/22(empId) works fine. employee/edit/22 works too but update does not work.
I am not using any assigned sequence. Just some random integer. Maybe a better example would be to use SSN instead of empId.
Thank you once again.
You are a great help buddy. Appreciate your time and patience. I am not writing any special update or save (I am too new to dig too deep). Just using grails to generate-all. However, I did find a workaround. I changed the domain class to add variable id (Long) and added empId setter method to allocate the empId value to id. That did it. Here is my code.
class Employee {
Long id
Long empId
String firstName
String lastName
static constraints = {
empId()
firstName()
lastName()
}
static mapping = {
id generator:'assigned', name:'empId', column: 'emp_id'
version false
}
public void setEmpId(Long empId){
this.empId = empId
this.id = empId
}
}
Please feel free to suggest if you have a better way of doing that.
Regards
Rocky
If you are using the "assigned" sequence, then you have to assign the objects ids yourself before saving them. Otherwise your objects will be saved with a null or 'default 0' id. If you want GORM to assign an id for you, you need to use another type of generator, like "sequence" generator. It would be like:
id name: 'customId', generator: 'sequence', params: [sequence:'some_sequence']
More info on id generators here.
Ok, I wrote this question up earlier today but I decided to delete it because I thought the question wasn't worded very well. I decided to wait until I had more time to compose it at home :).
I am just getting started with IOC/DI. I have done some research on which framework to use and decided to give StructureMap a spin. The following is the first tutorial I used:
http://dimecasts.net/Casts/CastDetails/39 by Derik Whittaker.
Anyways, I got everything working like a dream with EVERYTHING is hosted in the same project.
Here is my sample code:
[PluginFamily("SMTest",IsSingleton=true)]
public interface IVehicle
{
byte TopSpeed {set;get;}
byte MPG { set; get; }
}
[Pluggable("SMTest")]
public class Car : IVehicle
{
private byte mTopSpeed;
private byte mMPG;
#region IVehicle Members
byte IVehicle.TopSpeed
{
get
{
return mTopSpeed;
}
set
{
mTopSpeed = value;
}
}
public interface IConsumer
{
IVehicle Car { get; set; }
}
[Pluggable("SMTest")]
public class Consumer : StructureMapBasic.IConsumer
{
private IVehicle mCar;
public Consumer(IVehicle lcar)
{
Car = lcar;
}
public IVehicle Car { set; get; }
byte IVehicle.MPG
{
get
{
return mMPG;
}
set
{
mMPG = value;
}
}
#endregion
}
So anyways if i create the project above into a command line program and do the following:
var consumer = ObjectFactory.GetInstance<IConsumer>();
It works perfectly. No problems at all. When i create a seperate project in the solution and then change the project above to a DLL. I get the following error:
Test method StructureMapBasic.ConsumerTest.ConsumerConstructorTest
threw exception: StructureMap.StructureMapException: StructureMap
Exception Code: 202 No Default Instance defined for PluginFamily
StructureMapBasic.IConsumer, StructureMapBasic, Version=1.0.0.0,
Culture=neutral, PublicKeyToken=null.
At first i thought maybe the StructureMap.Config file didn't get into the new projects bin folder but that wasn't the case. It was there. Everythign compiles just fine this problem happens at runtime. I'm sure the solution is very easy but for the life of me i can't figure out whats going wrong. Any help would be MUCH appreciated.
Thanks,
Ncage
I took me all friggn day to figure this out. At first i thought i was just being an idiot and i was missing something stupid. Well i was not. I thought it was related to different projects but it was not. I created a new console application that consumed my StructureMapped DLL (If just coined a term ;) ). Anyways, after trying to spend all day on this problem i FINALLY found a post that described the problem. its a freakn bug in MSTest (my project i was having problems with was created in MSTest). Xunit here i come. Here is a post that describes the issue from the same guy who created the tutorial video:
http://devlicio.us/blogs/derik_whittaker/archive/2008/07/23/mstest-why-i-hate-you-you-cause-me-too-much-friction.aspx