We Keep Coding

More informative comparison of objects in C#

In my C# testing, I often want to compare two objects of the same type (typically an expected object against the actual object), but I want to allow for some flexibility. For example, there may be timestamp fields that I know can't be equal or some fields that I just want to ignore when comparing the objects.
Most importantly, I want to provide an informative message that describes where the two object properties' values differ in order that I can quickly identify what the problem is. For example, a message that says "Source property Name value Fred does not match target property Name value Freda".
The standard Equals and Comparer methods just seem to return ints or Booleans which don't provide enough information for me. At the moment, my object comparison methods return a custom type that has two fields (a boolean and a message), but my thinking is that there must be a more standard way to do this. These days, perhaps a Tuple might be the way to go, but I would welcome suggestions.

"Comparison" might not be the word for what you're trying to do. That word already has a common meaning in this context. We compare objects for equality, which returns a boolean - they are equal or they are not. Or we compare them to see which is greater. That returns an int which can indicate that one or the other is greater, or that they are equal. This is helpful when sorting objects.
What you're trying to do is determine specific differences between objects. I wouldn't try to write something generic that handles different types of objects unless you intend for them to be extremely simple. That gets really complicated as you get into properties that return additional complex objects or collections or collections of complex objects. It's not impossible, just rarely worth the effort compared to just writing a method that compares the particular type you want to compare.
Here's a few interfaces and classes that could make the task a little easier and more consistent. But to be honest it's hard to tell what to do with this. And again, it gets complicated if you're dealing with nested complex properties. What happens if two properties both contain lists of some other object, and all the items in those lists are the same except one on each side that have a differing property. Or what if they're all different? In that case how would you describe the "inequality" of the parent objects? It might be useful to know that they are or are not equal, but less so to somehow describe the difference.
public interface IInstanceComparer<T>
{
IEnumerable<PropertyDifference> GetDifferences(T left, T right);
}
public abstract class InstanceComparer<T> : IInstanceComparer<T>
{
public IEnumerable<PropertyDifference> GetDifferences(T left, T right)
{
var result = new List<PropertyDifference>();
PopulateDifferences(left, right, result);
return result;
}
public abstract void PopulateDifferences(T left, T right,
List<PropertyDifference> differences);
}
public class PropertyDifference
{
public PropertyDifference(string propertyName, string leftValue,
string rightValue)
{
PropertyName = propertyName;
LeftValue = leftValue;
RightValue = rightValue;
}
public string PropertyName { get; }
public string LeftValue { get; }
public string RightValue { get; }
}
public class Animal
{
public string Name { get; }
public int NumberOfLimbs { get; }
public DateTime Created { get; }
}
public class AnimalDifferenceComparer : InstanceComparer<Animal>
{
public override void PopulateDifferences(Animal left, Animal right,
List<PropertyDifference> differences)
{
if(left.Name != right.Name)
differences.Add(new PropertyDifference("Name", left.Name, right.Name));
if(left.NumberOfLimbs!=right.NumberOfLimbs)
differences.Add(new PropertyDifference("NumberOfLimbs",
left.NumberOfLimbs.ToString(),
right.NumberOfLimbs.ToString()));
}
}

You could use extension methods to do this. For example:
public static Extensions
{
public static void CompareWithExpected(this <type> value, <type> expected)
{
Assert.AreEqual(expected.Property1, value.Property1, "Property1 did not match expected";
Assert.AreEqual(expected.Property2, value.Property2, "Property2 did not match expected";
}
}
Then this can be used as follows:
public void TestMethod()
{
// Arrange
...
// Act
...
// Assert
value.CompareWithExpected(expected);
}
You could have any number of these extension methods allowing you the flexibility to check only certain values etc.
This also means you do not need to pollute your types with what is essentially test code.

Effects of returning a self reference from an instance method in C#

Let's say I have an interface called IConvertableModel and it helps me to convert some MVC models to/from DTO objects as shown below:
public class DisplayEditModel : IConvertableModel<Display>
{
[HiddenInput(DisplayValue = false)]
public int ObjectId { get; set; }
[StringLength(255)]
public string Description { get; set; }
public Display ToDto()
{
return new Display
{
Description = Description,
ObjectId = ObjectId,
};
}
public void SetFromDto(Display dto)
{
Description = dto.Description;
ObjectId = dto.ObjectId;
}
}
But there is one problem with this approach and that is it doesn't allow me do this :
var dto = _dtoRepository.GetFirstDto();
return new DisplayEditModel().SetFromDto(dto);
Instead I should do the following:
var dto = _dtoRepository.GetFirstDto();
var model = new DisplayEditModel();
model.SetFromDto(dto);
return model;
and this is adding extra two lines of code and little bit complexity in the long run.
What I am thinking is to convert SetFromDto method to something like this:
public DisplayEditModel SetFromDto(Display dto)
{
Description = dto.Description;
ObjectId = dto.ObjectId;
return this;
}
I think the benefit of this code is obvious but I also like to learn whether this harms code readability and leads to unexpected results for developers in the long run and if you think anything else, what would you recommend.
Note: Because of the interfaces reasons, I am not thinking to implement a constructor method.

A few thoughts, to begin with:
Adding lines of code is not the same as adding complexity. Having three statements, where each does a simple operation, is not necessarily harder to maintain or understand than a single statement with three operations inside of it.
When a method that begins with Set..., programmers will automatically assume that some stateful values of the target object are going to be changed by this method. It is rare for Set methods to have a return value. Property setters in C# actually "return" the original value passed into them, so you can chain setters:
int i = foo.A = 2;
So the precedent is generally against returning "this" from a set method specifically.
Chaining in general is most useful/desired when you're expecting several operations to be performed, one after the other. For example, C# provides nice initialization syntax so you can "chain" a bunch of different property setters on the same object:
var foo = new Foo { A = 1, B = 2 };
You can see how chaining is fulfilling the need to perform similar, grouped, repetitive operations that typically get performed all together. That is not the problem that you are trying to solve.
If your main gripe is that you don't like having three lines of code, why not just use a helper whose name indicates what you're trying to do?
TModel MapToModel<TModel, TDto>(TDto dto, TModel model)
where TModel : IConvertableModel<TDto>
{
model.SetFromDto(dto);
return model;
}
// usage:
var dto = _dtoRepository.GetFirstDto();
return MapToModel(dto, new DisplayEditModel());
... or even:
TModel CreateModel<TModel, TDto>(TDto dto)
where TModel : IConvertableModel<TDto>, new()
{
var model = new TModel();
return MapToModel(dto, model);
}
// usage:
var dto = _dtoRepository.GetFirstDto();
return CreateModel<DisplayEditModel>(dto);
This is simple, readable, and feasible, whereas the approach you're suggesting would break the IConvertableModel<Display> interface:
public interface IConvertableModel<TDto>
{
public TDto ToDto();
public ??? SetFromDto(TDto dto);
}
What would SetFromDto return? You would have to define another generic type on IConvertableModel.
public interface IConvertableModel<TDto, TModel> {
public TDto ToDto();
public TModel SetFromDto(TDto dto);
}
But this wouldn't really indicate that the SetFromDto method is necessarily returning itself, because it allows for a class that is not a TModel to implement IConvertableModel to convert between two other types.
Now, you could go out of your way to push the generics even farther:
public interface IConvertableModel<TDto, TModel>
where TModel : IConvertableModel<TDto, TModel>
{...}
But this still allows for some fudging, and the interface cannot guarantee that you are really returning "this" object. All in all, I'm not a big fan of that approach.

Rather than having DisplayEditModel have a get/set method for a Display object to get/set the values, just use a property that doesn't actually have a separate backing store:
public Display Display
{
get
{
return new Display
{
Description = Description,
ObjectId = ObjectId,
};
}
set
{
Description = value.Description;
ObjectId = value.ObjectId;
}
}
Now you can use an object initializer with this property when creating a model:
return new DisplayEditModel() { Display = dto };

This is a very javascript way of approaching this problem, though it has it's benefits. In the context of C#, it is a little bit strange though libraries such as LINQ do this to allow chaining together function calls.
My only worry about this, is that this has to be a class that does this consistently. Implementing a chaining function return pattern is not really a convenience as much as it is a design choice. The rule to follow in this case, would be to return this every time you mutate the object.
Chaining also may not be worth it performance wise. Something that can be done by wrapping all those operations into a single function is much faster. For instance:
MyVector.setX(1).SetY(1).SetZ(1).SetW(0)
is a lot slower than simply
MyVector.set(1, 1, 1, 0)
because now you are now doing excessive stack operations to do something fairly simple. It only becomes worth it on very large operations that take up the bulk of the computing time and make sense to chain together. For this reason, LINQ allows you to chain things together.
I wouldn't say that it necessary "harms" or is dangerous. We are in the world of a managed language, so we don't have direct access to that memory location (unlike C/C++). So I would just call it a design choice which can be fairly powerful in some cases and not so much in others.

As noted, chainable methods work fine but are not as common in C# as in some other languages. If the extra lines of code only happen in one place, I'd just leave it alone. If it's really bugging you or you do it a lot, then consider implementing a special constructor for it:
public void DisplayEditModel(Display dto)
{
this.SetFrom(dto);
}
or a static factory method:
public static DisplayEditModel CreateFrom(Display dto)
{
var model = new DisplayEditModel();
model.SetFrom(dto);
return model;
}
Either option has a clear intent, lets you create and return the object in a single line, and is idiomatic. It does require a few extra lines of code in DisplayEditModel, but I doubt it will be a serious problem.

A dictionary with a built-in factory?

I have a solution that works, but for educational purposes I want to understand if there is a better/cleaner/right way to do it.
Problem: In my "client" app I have a dictionary Dictionary<String, PremiseObject> where the key (String) is a immutable URL to a resource (it's actually a REST URL). PremiseObject is the base type of a whole set of derived classes; thus the Dictionary actually contains a family of classes all derived from PremiseObject.
A key requirement is I want to try to 'guarantee' that no PremiseObjects get created OUTSIDE of the dictionary.
Solution: I have the following function for getting a object out of the dictionary. It either accesses the existing instance, or if it does not exist creates it:
public PremiseObject GetOrCreateServerObject(string premiseObjectType, string location)
{
PremiseObject po = null;
if (!premiseObjects.TryGetValue(location, out po))
{
string classname;
// Create an instance of the right PremiseObject derived class:
po = // gobbly-gook that is not relevant to this question.
premiseObjects.Add(location, po);
}
else
{
Debug.WriteLine("Already exists: {0}", location);
}
return po;
}
Callers do this:
DoorSensor door =
(DoorSensor)server.GetOrCreateServerObject("DoorSensor",
"http://xyz/FrontDoor");
Works great. But I think there's a pattern or design that would elegantly allow me to encapsulate the "single-instance of each object contained in the dictionary" more.
For example, callers could do this:
DoorSensor door = null;
if (!server.ServerObjects.TryGetValue("DoorSensor",
"http://xyz/FrontDoor",
out door))
Debug.WriteLine("Something went very wrong");
I'm not really what to call this pattern. My ServerObjects are "single-instanced" by location. And my GetOrCreateServerObject is like a factory that lazy creates.
But it's possible for instances to be created that don't get put into the dictionary, which could lead to problems.
Like I said, what I have works... Cheers!
UPDATE 1/26/2011 10:13PM -
I just realized a potential problem: On the server side the object represented by a location/URL can actually be multi-inherited. It is THEORETICALLY possible for an object to be both a DoorSensor and an DigitalRelay.
I currently don't care about any of those cases (e.g. for garage doors I simplified my example above; there really is no DoorSensor I exposed, just a GarageDoorOpener which includes BOTH properties for sensing (e.g. Status) and actuation (e.g. Trigger). But this puts a wrinkle in my whole scheme if I were to care. Since this project is just for me :-) I am going to declare I don't care and document it.

I would propose the following simple idea:
PremiseObject's constructor is declared internal.
A special factory object is responsible for creating (or returning an already created) instances. The dictionary is a part of the factory.
Clients are located in another assembly.
This way PremiseObjects can be created by clients only through the factory. This way you can guarantee that only single instance of object exists for each location.
A variant of the idea would be to declare the PremiseObject's constructor private, and declare the factory a friend; but (unlike C++) C# doesn't have a friend notion.

Ok you can probably avoid a parameter and a cast (in the consumer code any way) with a generic method.
public abstract class PremiseObject
{
protected PremiseObject()
{
}
public string Location { get; set; }
public static void GetSensor<T>(string location, out T sensor)
where T : PremiseObject, new()
{
PremiseObject so;
if(_locationSingltons.TryGetValue(location, out so))
{
sensor = (T) so; // this will throw and exception if the
// wrong type has been created.
return;
}
sensor = new T();
sensor.Location = location;
_locationSingltons.Add(location, sensor);
}
private static Dictionary<string, PremiseObject> _locationSingltons
= new Dictionary<string, PremiseObject>();
}
Then the calling code looks a bit nicer:
DoorSensor frontDoor;
PremiseObject.GetSensor("http://FrontDoor/etc", out frontDoor);
So I like that calling convention - if you want to stay away from throwing an exception you can change the return type to bool and indicate failure that way. Personally I wouls say that an exception is what you want.
You may prefer the call without the out parameter - but if you do that then you have to supply the type to the method call - anyway defining the factory method would look like this:
public static T GetSensor<T>(string location) where T : PremiseObject, new()
{
PremiseObject so;
if (_locationSingltons.TryGetValue(location, out so))
{
return (T)so; // this will throw and exception if the
// wrong type has been created.
}
T result = new T();
result.Location = location;
_locationSingltons.Add(location, result);
return result;
}
Then the calling code looks like this:
var frontDoor2 = PremiseObject.GetSensor<DoorSensor>("http://FrontDoor/etc");
I like both these approaches because nothing has to be repeated. The type of the PremiseObject only gets stated once - there is no need for a string defining the type.

If you want to be really, really sure that no instances of PremiseObject get created that aren't placed in the dictionary, you could make the constructors all private, and create a static constructor (for each subclass) that took as a parameter the Dictionary object you're referring to. This static constructor would check the dictionary object to make sure that there wasn't an existing instance, and then return either the new or the existing instance as required. So something like this:
public class PremiseObject
{
public static Dictionary<string, PremiseObject> PremiseObjects { get; private set; }
static PremiseObject()
{
PremiseObjects = new Dictionary<string, PremiseObject>();
}
}
public class DerivedPremiseObject : PremiseObject
{
private DerivedPremiseObject()
{
}
public static DerivedPremiseObject GetDerivedPremiseObject(string location)
{
DerivedPremiseObject po = null;
if (!PremiseObject.PremiseObjects.TryGetValue(location, out po))
{
po = new DerivedPremiseObject();
PremiseObject.PremiseObjects.Add(location, po);
}
return po;
}
}
And there are a variety of similar strategies you could use. The key is to somehow make the constructor private and only allow access to the constructor through a static method that enforces the logic of the class construction.

Perhaps you could make PremiseObject a singleton, then you wouldn't have to worry about each object in the dictionary beign a single instance?

In the general case, setting access modifiers on your constructors should do the trick of not allowing anyone external to create the objects (barring reflection). However, these would need to be internal, so anything else in the assembly would be able to instantiate them.
I suspect many of your requirements may be met by using an off the shelf dependency injection container that supports singleton instances. It feels close, but maybe not quite the same. (possibly StrutureMap, Ninject, Castle Windsor, or Unity in no particular order)

C# design for an object where some properties are expensive: excuse to make it mutable?

Yes, I know, yet another question about mutable objects. See this for general background and this for the closest analogue to my question. (though it has some C++ specific overtones that don't apply here)
Let's assume that the following pseudo code represents the best interface design. That is, it's the clearest expression of the business semantics (as they stand today) into OO type. Naturally, the UglyData and the things we're tasked to do with it are subject to incremental change.
public class FriendlyWrapper
{
public FriendlyWrapper(UglyDatum u)
{
Foo = u.asdf[0].f[0].o.o;
Bar = u.barbarbar.ToDooDad();
Baz = u.uglyNameForBaz;
// etc
}
public Widget Foo { get; private set; }
public DooDad Bar { get; private set; }
public DooDad Baz { get; private set; }
// etc
public WhizBang Expensive1 { get; private set; }
public WhizBang Expensive2 { get; private set; }
public void Calculate()
{
Expensive1 = Calc(Foo, Bar);
Expensive2 = Calc(Foo, Baz);
}
private WhizBang Calc(Widget a, DooDad b) { /* stuff */ }
public override void ToString()
{
return string.Format("{0}{1}{2}{3}{4}", Foo, Bar, Baz, Expensive1 ?? "", Expensive2 ?? "");
}
}
// Consumer 1 is happy to work with just the basic wrapped properties
public string Summarize()
{
var myStuff = from u in data
where IsWhatIWant(u)
select new FriendlyWrapper(u);
var sb = new StringBuilder();
foreach (var s in myStuff)
{
sb.AppendLine(s.ToString());
}
return sb.ToString();
}
// Consumer 2's job is to take the performance hit up front. His callers might do things
// with expensive properties (eg bind one to a UI element) that should not take noticeable time.
public IEnumerable<FriendlyWrapper> FetchAllData(Predicate<UglyDatum> pred)
{
var myStuff = from u in data
where pred(u)
select new FriendlyWrapper(u);
foreach (var s in myStuff)
{
s.Calculate(); // as written, this doesn't do what you intend...
}
return myStuff;
}
What's the best route here? Options I can see:
Mutable object with an explicit Calculate() method, as above
Mutable object where expensive calculations are done in the getters (and probably cached)
Split into two objects where one inherits (or perhaps composes?) from the other
Some sort of static + locking mechanism, as in the C++ question linked above
I'm leaning toward #2 myself. But every route has potential pitfalls.
If you choose #1 or #2, then how would you implement Consumer2's loop over mutables in a clear, correct manner?
If you choose #1 or #3, how would you handle future situations where you only want to calculate some properties but not others? Willing to create N helper methods / derived classes?
If you choose #4, I think you're crazy, but feel free to explain

In your case, since you're using LINQ, you're only going to constructing these objects in cases where you want the calculation.
If that is your standard usage pattern, I would just put the expensive calculation directly in the constructor. Using lazy initialization is always slower unless you plan to have some cases where you do not calculate. Doing the calculation in the getters will not save anything (at least in this specific case).
As for mutability - mutable objects with reference syntax and identity (ie: classes in C#) are really okay - it's more a problem when you're dealing with mutable value types (ie: structs). There are many, many mutable classes in the .NET BCL - and they don't cause issues. The problem is typically more of one when you start dealing with value types. Mutable value types lead to very unexpected behavior.
In general, I'd turn this question upside down - How and where are you going to use this object? How can you make this object the most performant (if it's been determined to be problematic) without affecting usability? Your 1), 3) and 4) options would all make usability suffer, so I'd avoid them. In this case, doing 2) won't help. I'd just put it in the constructor, so your object's always in a valid state (which is very good for usability and maintainability).

C# return a variable as read only from get; set;

I swear I have seen an example of this but have been googling for a bit and can not find it.
I have a class that has a reference to an object and need to have a GET; method for it. My problem is that I do not want anyone to be able to fiddle with it, i.e. I want them to get a read only version of it, (note I need to be able to alter it from within my class).
Thanks

No, there's no way of doing this. For instance, if you return a List<string> (and it's not immutable) then callers will be able to add entries.
The normal way round this is to return an immutable wrapper, e.g. ReadOnlyCollection<T>.
For other mutable types, you may need to clone the value before returning it.
Note that just returning an immutable interface view (e.g. returning IEnumerable<T> instead of List<T>) won't stop a caller from casting back to the mutable type and mutating.
EDIT: Note that apart from anything else, this kind of concern is one of the reasons why immutable types make it easier to reason about code :)

Return a reference to a stripped-down interface:
interface IFoo
string Bar { get; }
class ClassWithGet
public IFoo GetFoo(...);

If the object isn't too complicated/extensive then write an wrapper around it.
for example:
class A {
public string strField = 'string';
public int intField = 10;
}
class AWrapper {
private A _aObj;
public AWrapper(A aobj) {
_aObj = A;
}
public string strField {
get {
return _aObj.strField;
}
}
public int intField {
get {
return _aObj.intField;
}
}
}
So now all you do is give your client code an instance of the AWrapper class so that they may only use what you allow them to see.
this may get a bit complicated and may not scale well if your base class is not set in stone, but for most simple situation it may just do the trick. I think this is called a facade pattern(but don't quote me on that =) )

This isn't possible. Get and set accessors to reference types get and set the reference to the object. You can prevent changes to the reference by using a private (or internal) setter, but you cannot prevent changes to the object itself if it's exposed by a getter.

Your question reads like you're looking for:
public PropertyName { get; private set; }
But then, given the answers so far I'm not sure I'm interpreting your question correctly. Besides, who am I to question Jon Skeet? :)

i agree with ReadOnlyCollection
See my simple code:
private List<Device> _devices;
public readonly System.Collections.ObjectModel.ReadOnlyCollection<Device> Devices
{
get
{
return (_devices.AsReadOnly());
}
}
ReadOnlyCollection dosen't has Add method so user cant add properties to it.BUT ther is no warranty that if user can modify objects by calling their methods....

I have faced this problem in a certain way.
I have a CategoryViewModel class, which have a property Category that I want private read-only :
public CategoryViewModel
{
private Category { get; }
}
In fact, I want it to be exported as read-only to other class. However I can't do such thing.
In my case (maybe it will help some other guys), I want to add it to a repository. The only way that I've found is to have a function with the repository as param 1, and an Action as param 2 :
public void ApplyAction(ICategoryRepository repo, Action<ICategoryRepository, Category> action)
{
action(repo, Category);
}
Like that, from elsewhere, I can do such thing :
categoryViewModel.ApplyAction(_repository, (r, c) => r.MarkForInsertOrUpdate(c));
This can help other to expose there property only for certains cases and can manage them.