Comparing object properties in c# [closed] - c#

Closed. This question is opinion-based. It is not currently accepting answers.
Closed 4 years ago.
Locked. This question and its answers are locked because the question is off-topic but has historical significance. It is not currently accepting new answers or interactions.
This is what I've come up with as a method on a class inherited by many of my other classes. The idea is that it allows the simple comparison between properties of Objects of the same Type.
Now, this does work - but in the interest of improving the quality of my code I thought I'd throw it out for scrutiny. How can it be better/more efficient/etc.?
/// <summary>
/// Compare property values (as strings)
/// </summary>
/// <param name="obj"></param>
/// <returns></returns>
public bool PropertiesEqual(object comparisonObject)
{
Type sourceType = this.GetType();
Type destinationType = comparisonObject.GetType();
if (sourceType == destinationType)
{
PropertyInfo[] sourceProperties = sourceType.GetProperties();
foreach (PropertyInfo pi in sourceProperties)
{
if ((sourceType.GetProperty(pi.Name).GetValue(this, null) == null && destinationType.GetProperty(pi.Name).GetValue(comparisonObject, null) == null))
{
// if both are null, don't try to compare (throws exception)
}
else if (!(sourceType.GetProperty(pi.Name).GetValue(this, null).ToString() == destinationType.GetProperty(pi.Name).GetValue(comparisonObject, null).ToString()))
{
// only need one property to be different to fail Equals.
return false;
}
}
}
else
{
throw new ArgumentException("Comparison object must be of the same type.","comparisonObject");
}
return true;
}

I was looking for a snippet of code that would do something similar to help with writing unit test. Here is what I ended up using.
public static bool PublicInstancePropertiesEqual<T>(T self, T to, params string[] ignore) where T : class
{
if (self != null && to != null)
{
Type type = typeof(T);
List<string> ignoreList = new List<string>(ignore);
foreach (System.Reflection.PropertyInfo pi in type.GetProperties(System.Reflection.BindingFlags.Public | System.Reflection.BindingFlags.Instance))
{
if (!ignoreList.Contains(pi.Name))
{
object selfValue = type.GetProperty(pi.Name).GetValue(self, null);
object toValue = type.GetProperty(pi.Name).GetValue(to, null);
if (selfValue != toValue && (selfValue == null || !selfValue.Equals(toValue)))
{
return false;
}
}
}
return true;
}
return self == to;
}
EDIT:
Same code as above but uses LINQ and Extension methods :
public static bool PublicInstancePropertiesEqual<T>(this T self, T to, params string[] ignore) where T : class
{
if (self != null && to != null)
{
var type = typeof(T);
var ignoreList = new List<string>(ignore);
var unequalProperties =
from pi in type.GetProperties(BindingFlags.Public | BindingFlags.Instance)
where !ignoreList.Contains(pi.Name) && pi.GetUnderlyingType().IsSimpleType() && pi.GetIndexParameters().Length == 0
let selfValue = type.GetProperty(pi.Name).GetValue(self, null)
let toValue = type.GetProperty(pi.Name).GetValue(to, null)
where selfValue != toValue && (selfValue == null || !selfValue.Equals(toValue))
select selfValue;
return !unequalProperties.Any();
}
return self == to;
}
public static class TypeExtensions
{
/// <summary>
/// Determine whether a type is simple (String, Decimal, DateTime, etc)
/// or complex (i.e. custom class with public properties and methods).
/// </summary>
/// <see cref="http://stackoverflow.com/questions/2442534/how-to-test-if-type-is-primitive"/>
public static bool IsSimpleType(
this Type type)
{
return
type.IsValueType ||
type.IsPrimitive ||
new[]
{
typeof(String),
typeof(Decimal),
typeof(DateTime),
typeof(DateTimeOffset),
typeof(TimeSpan),
typeof(Guid)
}.Contains(type) ||
(Convert.GetTypeCode(type) != TypeCode.Object);
}
public static Type GetUnderlyingType(this MemberInfo member)
{
switch (member.MemberType)
{
case MemberTypes.Event:
return ((EventInfo)member).EventHandlerType;
case MemberTypes.Field:
return ((FieldInfo)member).FieldType;
case MemberTypes.Method:
return ((MethodInfo)member).ReturnType;
case MemberTypes.Property:
return ((PropertyInfo)member).PropertyType;
default:
throw new ArgumentException
(
"Input MemberInfo must be if type EventInfo, FieldInfo, MethodInfo, or PropertyInfo"
);
}
}
}

UPDATE: The latest version of Compare-Net-Objects is located on GitHub , has NuGet package and Tutorial. It can be called like
//This is the comparison class
CompareLogic compareLogic = new CompareLogic();
ComparisonResult result = compareLogic.Compare(person1, person2);
//These will be different, write out the differences
if (!result.AreEqual)
Console.WriteLine(result.DifferencesString);
Or if you need to change some configuration, use
CompareLogic basicComparison = new CompareLogic()
{ Config = new ComparisonConfig()
{ MaxDifferences = propertyCount
//add other configurations
}
};
Full list of configurable parameters is in ComparisonConfig.cs
Original answer:
The limitations I see in your code:
The biggest one is that it doesn't do a deep object comparison.
It doesn't do an element by element comparison in case properties are lists or contain lists as elements (this can go n-levels).
It doesn't take into account that some type of properties should not be compared (e.g. a Func property used for filtering purposes, like the one in the PagedCollectionView class).
It doesn't keep track of what properties actually were different (so you can show in your assertions).
I was looking today for some solution for unit-testing purposes to do property by property deep comparison and I ended up using: http://comparenetobjects.codeplex.com.
It is a free library with just one class which you can simply use like this:
var compareObjects = new CompareObjects()
{
CompareChildren = true, //this turns deep compare one, otherwise it's shallow
CompareFields = false,
CompareReadOnly = true,
ComparePrivateFields = false,
ComparePrivateProperties = false,
CompareProperties = true,
MaxDifferences = 1,
ElementsToIgnore = new List<string>() { "Filter" }
};
Assert.IsTrue(
compareObjects.Compare(objectA, objectB),
compareObjects.DifferencesString
);
Also, it can be easily re-compiled for Silverlight. Just copy the one class into a Silverlight project and remove one or two lines of code for comparisons that are not available in Silverlight, like private members comparison.

I think it would be best to follow the pattern for Override Object#Equals()
For a better description: Read Bill Wagner's Effective C# - Item 9 I think
public override Equals(object obOther)
{
if (null == obOther)
return false;
if (object.ReferenceEquals(this, obOther)
return true;
if (this.GetType() != obOther.GetType())
return false;
# private method to compare members.
return CompareMembers(this, obOther as ThisClass);
}
Also in methods that check for equality, you should return either true or false. either they are equal or they are not.. instead of throwing an exception, return false.
I'd consider overriding Object#Equals.
Even though you must have considered this, using Reflection to compare properties is supposedly slow (I dont have numbers to back this up). This is the default behavior for valueType#Equals in C# and it is recommended that you override Equals for value types and do a member wise compare for performance. (Earlier I speed-read this as you have a collection of custom Property objects... my bad.)
Update-Dec 2011:
Of course, if the type already has a production Equals() then you need another approach.
If you're using this to compare immutable data structures exclusively for test purposes, you shouldn't add an Equals to production classes (Someone might hose the tests by chainging the Equals implementation or you may prevent creation of a production-required Equals implementation).

If performance doesn't matter, you could serialize them and compare the results:
var serializer = new XmlSerializer(typeof(TheObjectType));
StringWriter serialized1 = new StringWriter(), serialized2 = new StringWriter();
serializer.Serialize(serialized1, obj1);
serializer.Serialize(serialized2, obj2);
bool areEqual = serialized1.ToString() == serialized2.ToString();

I think the answer of Big T was quite good but the deep comparison was missing, so I tweaked it a little bit:
using System.Collections.Generic;
using System.Reflection;
/// <summary>Comparison class.</summary>
public static class Compare
{
/// <summary>Compare the public instance properties. Uses deep comparison.</summary>
/// <param name="self">The reference object.</param>
/// <param name="to">The object to compare.</param>
/// <param name="ignore">Ignore property with name.</param>
/// <typeparam name="T">Type of objects.</typeparam>
/// <returns><see cref="bool">True</see> if both objects are equal, else <see cref="bool">false</see>.</returns>
public static bool PublicInstancePropertiesEqual<T>(T self, T to, params string[] ignore) where T : class
{
if (self != null && to != null)
{
var type = self.GetType();
var ignoreList = new List<string>(ignore);
foreach (var pi in type.GetProperties(BindingFlags.Public | BindingFlags.Instance))
{
if (ignoreList.Contains(pi.Name))
{
continue;
}
var selfValue = type.GetProperty(pi.Name).GetValue(self, null);
var toValue = type.GetProperty(pi.Name).GetValue(to, null);
if (pi.PropertyType.IsClass && !pi.PropertyType.Module.ScopeName.Equals("CommonLanguageRuntimeLibrary"))
{
// Check of "CommonLanguageRuntimeLibrary" is needed because string is also a class
if (PublicInstancePropertiesEqual(selfValue, toValue, ignore))
{
continue;
}
return false;
}
if (selfValue != toValue && (selfValue == null || !selfValue.Equals(toValue)))
{
return false;
}
}
return true;
}
return self == to;
}
}

I would add the following line to the PublicInstancePropertiesEqual method to avoid copy & paste errors:
Assert.AreNotSame(self, to);

Do you override .ToString() on all of your objects that are in the properties? Otherwise, that second comparison could come back with null.
Also, in that second comparison, I'm on the fence about the construct of !( A == B) compared to (A != B), in terms of readability six months/two years from now. The line itself is pretty wide, which is ok if you've got a wide monitor, but might not print out very well. (nitpick)
Are all of your objects always using properties such that this code will work? Could there be some internal, non-propertied data that could be different from one object to another, but all exposed data is the same? I'm thinking of some data which could change over time, like two random number generators that happen to hit the same number at one point, but are going to produce two different sequences of information, or just any data that doesn't get exposed through the property interface.

If you are only comparing objects of the same type or further down the inheritance chain, why not specify the parameter as your base type, rather than object ?
Also do null checks on the parameter as well.
Furthermore I'd make use of 'var' just to make the code more readable (if its c#3 code)
Also, if the object has reference types as properties then you are just calling ToString() on them which doesn't really compare values. If ToString isn't overwridden then its just going to return the type name as a string which could return false-positives.

The first thing I would suggest would be to split up the actual comparison so that it's a bit more readable (I've also taken out the ToString() - is that needed?):
else {
object originalProperty = sourceType.GetProperty(pi.Name).GetValue(this, null);
object comparisonProperty = destinationType.GetProperty(pi.Name).GetValue(comparisonObject, null);
if (originalProperty != comparisonProperty)
return false;
The next suggestion would be to minimise the use of reflection as much as possible - it's really slow. I mean, really slow. If you are going to do this, I would suggest caching the property references. I'm not intimately familiar with the Reflection API, so if this is a bit off, just adjust to make it compile:
// elsewhere
Dictionary<object, Property[]> lookupDictionary = new Dictionary<object, Property[]>;
Property[] objectProperties = null;
if (lookupDictionary.ContainsKey(sourceType)) {
objectProperties = lookupProperties[sourceType];
} else {
// build array of Property references
PropertyInfo[] sourcePropertyInfos = sourceType.GetProperties();
Property[] sourceProperties = new Property[sourcePropertyInfos.length];
for (int i=0; i < sourcePropertyInfos.length; i++) {
sourceProperties[i] = sourceType.GetProperty(pi.Name);
}
// add to cache
objectProperties = sourceProperties;
lookupDictionary[object] = sourceProperties;
}
// loop through and compare against the instances
However, I have to say that I agree with the other posters. This smells lazy and inefficient. You should be implementing IComparable instead :-).

here is revised one to treat null = null as equal
private bool PublicInstancePropertiesEqual<T>(T self, T to, params string[] ignore) where T : class
{
if (self != null && to != null)
{
Type type = typeof(T);
List<string> ignoreList = new List<string>(ignore);
foreach (PropertyInfo pi in type.GetProperties(BindingFlags.Public | BindingFlags.Instance))
{
if (!ignoreList.Contains(pi.Name))
{
object selfValue = type.GetProperty(pi.Name).GetValue(self, null);
object toValue = type.GetProperty(pi.Name).GetValue(to, null);
if (selfValue != null)
{
if (!selfValue.Equals(toValue))
return false;
}
else if (toValue != null)
return false;
}
}
return true;
}
return self == to;
}

I ended up doing this:
public static string ToStringNullSafe(this object obj)
{
return obj != null ? obj.ToString() : String.Empty;
}
public static bool Compare<T>(T a, T b)
{
int count = a.GetType().GetProperties().Count();
string aa, bb;
for (int i = 0; i < count; i++)
{
aa = a.GetType().GetProperties()[i].GetValue(a, null).ToStringNullSafe();
bb = b.GetType().GetProperties()[i].GetValue(b, null).ToStringNullSafe();
if (aa != bb)
{
return false;
}
}
return true;
}
Usage:
if (Compare<ObjectType>(a, b))
Update
If you want to ignore some properties by name:
public static string ToStringNullSafe(this object obj)
{
return obj != null ? obj.ToString() : String.Empty;
}
public static bool Compare<T>(T a, T b, params string[] ignore)
{
int count = a.GetType().GetProperties().Count();
string aa, bb;
for (int i = 0; i < count; i++)
{
aa = a.GetType().GetProperties()[i].GetValue(a, null).ToStringNullSafe();
bb = b.GetType().GetProperties()[i].GetValue(b, null).ToStringNullSafe();
if (aa != bb && ignore.Where(x => x == a.GetType().GetProperties()[i].Name).Count() == 0)
{
return false;
}
}
return true;
}
Usage:
if (MyFunction.Compare<ObjType>(a, b, "Id","AnotherProp"))

You can optimize your code by calling GetProperties only once per type:
public static string ToStringNullSafe(this object obj)
{
return obj != null ? obj.ToString() : String.Empty;
}
public static bool Compare<T>(T a, T b, params string[] ignore)
{
var aProps = a.GetType().GetProperties();
var bProps = b.GetType().GetProperties();
int count = aProps.Count();
string aa, bb;
for (int i = 0; i < count; i++)
{
aa = aProps[i].GetValue(a, null).ToStringNullSafe();
bb = bProps[i].GetValue(b, null).ToStringNullSafe();
if (aa != bb && ignore.Where(x => x == aProps[i].Name).Count() == 0)
{
return false;
}
}
return true;
}

For completeness I want to add reference to
http://www.cyotek.com/blog/comparing-the-properties-of-two-objects-via-reflection
It has more complete logic than most of others answers on this page.
However I prefer Compare-Net-Objects library
https://github.com/GregFinzer/Compare-Net-Objects (referred by Liviu Trifoi's answer)
The library has NuGet package http://www.nuget.org/packages/CompareNETObjects and multiple options to configure.

Make sure objects aren't null.
Having obj1 and obj2:
if(obj1 == null )
{
return false;
}
return obj1.Equals( obj2 );

This works even if the objects are different. you could customize the methods in the utilities class maybe you want to compare private properties as well...
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
class ObjectA
{
public string PropertyA { get; set; }
public string PropertyB { get; set; }
public string PropertyC { get; set; }
public DateTime PropertyD { get; set; }
public string FieldA;
public DateTime FieldB;
}
class ObjectB
{
public string PropertyA { get; set; }
public string PropertyB { get; set; }
public string PropertyC { get; set; }
public DateTime PropertyD { get; set; }
public string FieldA;
public DateTime FieldB;
}
class Program
{
static void Main(string[] args)
{
// create two objects with same properties
ObjectA a = new ObjectA() { PropertyA = "test", PropertyB = "test2", PropertyC = "test3" };
ObjectB b = new ObjectB() { PropertyA = "test", PropertyB = "test2", PropertyC = "test3" };
// add fields to those objects
a.FieldA = "hello";
b.FieldA = "Something differnt";
if (a.ComparePropertiesTo(b))
{
Console.WriteLine("objects have the same properties");
}
else
{
Console.WriteLine("objects have diferent properties!");
}
if (a.CompareFieldsTo(b))
{
Console.WriteLine("objects have the same Fields");
}
else
{
Console.WriteLine("objects have diferent Fields!");
}
Console.Read();
}
}
public static class Utilities
{
public static bool ComparePropertiesTo(this Object a, Object b)
{
System.Reflection.PropertyInfo[] properties = a.GetType().GetProperties(); // get all the properties of object a
foreach (var property in properties)
{
var propertyName = property.Name;
var aValue = a.GetType().GetProperty(propertyName).GetValue(a, null);
object bValue;
try // try to get the same property from object b. maybe that property does
// not exist!
{
bValue = b.GetType().GetProperty(propertyName).GetValue(b, null);
}
catch
{
return false;
}
if (aValue == null && bValue == null)
continue;
if (aValue == null && bValue != null)
return false;
if (aValue != null && bValue == null)
return false;
// if properties do not match return false
if (aValue.GetHashCode() != bValue.GetHashCode())
{
return false;
}
}
return true;
}
public static bool CompareFieldsTo(this Object a, Object b)
{
System.Reflection.FieldInfo[] fields = a.GetType().GetFields(); // get all the properties of object a
foreach (var field in fields)
{
var fieldName = field.Name;
var aValue = a.GetType().GetField(fieldName).GetValue(a);
object bValue;
try // try to get the same property from object b. maybe that property does
// not exist!
{
bValue = b.GetType().GetField(fieldName).GetValue(b);
}
catch
{
return false;
}
if (aValue == null && bValue == null)
continue;
if (aValue == null && bValue != null)
return false;
if (aValue != null && bValue == null)
return false;
// if properties do not match return false
if (aValue.GetHashCode() != bValue.GetHashCode())
{
return false;
}
}
return true;
}
}

Update on Liviu's answer above - CompareObjects.DifferencesString has been deprecated.
This works well in a unit test:
CompareLogic compareLogic = new CompareLogic();
ComparisonResult result = compareLogic.Compare(object1, object2);
Assert.IsTrue(result.AreEqual);

This method will get properties of the class and compare the values for each property. If any of the values are different, it will return false, else it will return true.
public static bool Compare<T>(T Object1, T object2)
{
//Get the type of the object
Type type = typeof(T);
//return false if any of the object is false
if (Object1 == null || object2 == null)
return false;
//Loop through each properties inside class and get values for the property from both the objects and compare
foreach (System.Reflection.PropertyInfo property in type.GetProperties())
{
if (property.Name != "ExtensionData")
{
string Object1Value = string.Empty;
string Object2Value = string.Empty;
if (type.GetProperty(property.Name).GetValue(Object1, null) != null)
Object1Value = type.GetProperty(property.Name).GetValue(Object1, null).ToString();
if (type.GetProperty(property.Name).GetValue(object2, null) != null)
Object2Value = type.GetProperty(property.Name).GetValue(object2, null).ToString();
if (Object1Value.Trim() != Object2Value.Trim())
{
return false;
}
}
}
return true;
}
Usage:
bool isEqual = Compare<Employee>(Object1, Object2)

To expand on #nawfal:s answer, I use this to test objects of different types in my unit tests to compare equal property names. In my case database entity and DTO.
Used like this in my test;
Assert.IsTrue(resultDto.PublicInstancePropertiesEqual(expectedEntity));
public static bool PublicInstancePropertiesEqual<T, Z>(this T self, Z to, params string[] ignore) where T : class
{
if (self != null && to != null)
{
var type = typeof(T);
var type2 = typeof(Z);
var ignoreList = new List<string>(ignore);
var unequalProperties =
from pi in type.GetProperties(BindingFlags.Public | BindingFlags.Instance)
where !ignoreList.Contains(pi.Name)
let selfValue = type.GetProperty(pi.Name).GetValue(self, null)
let toValue = type2.GetProperty(pi.Name).GetValue(to, null)
where selfValue != toValue && (selfValue == null || !selfValue.Equals(toValue))
select selfValue;
return !unequalProperties.Any();
}
return self == null && to == null;
}

sometimes you don't want to compare all public properties and want to compare only the subset of them, so in this case you can just move logic to compare the desired list of properties to abstract class
public abstract class ValueObject<T> where T : ValueObject<T>
{
protected abstract IEnumerable<object> GetAttributesToIncludeInEqualityCheck();
public override bool Equals(object other)
{
return Equals(other as T);
}
public bool Equals(T other)
{
if (other == null)
{
return false;
}
return GetAttributesToIncludeInEqualityCheck()
.SequenceEqual(other.GetAttributesToIncludeInEqualityCheck());
}
public static bool operator ==(ValueObject<T> left, ValueObject<T> right)
{
return Equals(left, right);
}
public static bool operator !=(ValueObject<T> left, ValueObject<T> right)
{
return !(left == right);
}
public override int GetHashCode()
{
int hash = 17;
foreach (var obj in this.GetAttributesToIncludeInEqualityCheck())
hash = hash * 31 + (obj == null ? 0 : obj.GetHashCode());
return hash;
}
}
and use this abstract class later to compare the objects
public class Meters : ValueObject<Meters>
{
...
protected decimal DistanceInMeters { get; private set; }
...
protected override IEnumerable<object> GetAttributesToIncludeInEqualityCheck()
{
return new List<Object> { DistanceInMeters };
}
}

my solution inspired from Aras Alenin answer above where I added one level of object comparison and a custom object for comparison results. I am also interested to get property name with object name:
public static IEnumerable<ObjectPropertyChanged> GetPublicSimplePropertiesChanged<T>(this T previous, T proposedChange,
string[] namesOfPropertiesToBeIgnored) where T : class
{
return GetPublicGenericPropertiesChanged(previous, proposedChange, namesOfPropertiesToBeIgnored, true, null, null);
}
public static IReadOnlyList<ObjectPropertyChanged> GetPublicGenericPropertiesChanged<T>(this T previous, T proposedChange,
string[] namesOfPropertiesToBeIgnored) where T : class
{
return GetPublicGenericPropertiesChanged(previous, proposedChange, namesOfPropertiesToBeIgnored, false, null, null);
}
/// <summary>
/// Gets the names of the public properties which values differs between first and second objects.
/// Considers 'simple' properties AND for complex properties without index, get the simple properties of the children objects.
/// </summary>
/// <typeparam name="T"></typeparam>
/// <param name="previous">The previous object.</param>
/// <param name="proposedChange">The second object which should be the new one.</param>
/// <param name="namesOfPropertiesToBeIgnored">The names of the properties to be ignored.</param>
/// <param name="simpleTypeOnly">if set to <c>true</c> consider simple types only.</param>
/// <param name="parentTypeString">The parent type string. Meant only for recursive call with simpleTypeOnly set to <c>true</c>.</param>
/// <param name="secondType">when calling recursively, the current type of T must be clearly defined here, as T will be more generic (using base class).</param>
/// <returns>
/// the names of the properties
/// </returns>
private static IReadOnlyList<ObjectPropertyChanged> GetPublicGenericPropertiesChanged<T>(this T previous, T proposedChange,
string[] namesOfPropertiesToBeIgnored, bool simpleTypeOnly, string parentTypeString, Type secondType) where T : class
{
List<ObjectPropertyChanged> propertiesChanged = new List<ObjectPropertyChanged>();
if (previous != null && proposedChange != null)
{
var type = secondType == null ? typeof(T) : secondType;
string typeStr = parentTypeString + type.Name + ".";
var ignoreList = namesOfPropertiesToBeIgnored.CreateList();
IEnumerable<IEnumerable<ObjectPropertyChanged>> genericPropertiesChanged =
from pi in type.GetProperties(BindingFlags.Public | BindingFlags.Instance)
where !ignoreList.Contains(pi.Name) && pi.GetIndexParameters().Length == 0
&& (!simpleTypeOnly || simpleTypeOnly && pi.PropertyType.IsSimpleType())
let firstValue = type.GetProperty(pi.Name).GetValue(previous, null)
let secondValue = type.GetProperty(pi.Name).GetValue(proposedChange, null)
where firstValue != secondValue && (firstValue == null || !firstValue.Equals(secondValue))
let subPropertiesChanged = simpleTypeOnly || pi.PropertyType.IsSimpleType()
? null
: GetPublicGenericPropertiesChanged(firstValue, secondValue, namesOfPropertiesToBeIgnored, true, typeStr, pi.PropertyType)
let objectPropertiesChanged = subPropertiesChanged != null && subPropertiesChanged.Count() > 0
? subPropertiesChanged
: (new ObjectPropertyChanged(proposedChange.ToString(), typeStr + pi.Name, firstValue.ToStringOrNull(), secondValue.ToStringOrNull())).CreateList()
select objectPropertiesChanged;
if (genericPropertiesChanged != null)
{ // get items from sub lists
genericPropertiesChanged.ForEach(a => propertiesChanged.AddRange(a));
}
}
return propertiesChanged;
}
Using the following class to store comparison results
[System.Serializable]
public class ObjectPropertyChanged
{
public ObjectPropertyChanged(string objectId, string propertyName, string previousValue, string changedValue)
{
ObjectId = objectId;
PropertyName = propertyName;
PreviousValue = previousValue;
ProposedChangedValue = changedValue;
}
public string ObjectId { get; set; }
public string PropertyName { get; set; }
public string PreviousValue { get; set; }
public string ProposedChangedValue { get; set; }
}
And a sample unit test:
[TestMethod()]
public void GetPublicGenericPropertiesChangedTest1()
{
// Define objects to test
Function func1 = new Function { Id = 1, Description = "func1" };
Function func2 = new Function { Id = 2, Description = "func2" };
FunctionAssignment funcAss1 = new FunctionAssignment
{
Function = func1,
Level = 1
};
FunctionAssignment funcAss2 = new FunctionAssignment
{
Function = func2,
Level = 2
};
// Main test: read properties changed
var propertiesChanged = Utils.GetPublicGenericPropertiesChanged(funcAss1, funcAss2, null);
Assert.IsNotNull(propertiesChanged);
Assert.IsTrue(propertiesChanged.Count == 3);
Assert.IsTrue(propertiesChanged[0].PropertyName == "FunctionAssignment.Function.Description");
Assert.IsTrue(propertiesChanged[1].PropertyName == "FunctionAssignment.Function.Id");
Assert.IsTrue(propertiesChanged[2].PropertyName == "FunctionAssignment.Level");
}

Related

Get nested type properties using reflection

As part of a test I am trying to compare similar objects using reflection.
The objects may or may not have multiple levels of nested params.
For example:
public class Connection{
public string Ip{get; set}
public string Id{get; set}
public string Key{get; set}
public string Transport{get; set}
public Parameters ParametersObj = new Parameters();
public class Parameters
{
public string AssignedName { get; set; }
public string CategoryType { get; set; }
public string Status { get; set; }
}
};
This is just an example of a class, I need this method to deal with any type of object without knowing the number of depth level.
I am doing something like this after I have made sure the two objects are of the same type.
bool result = true;
foreach (var objParam in firstObj.GetType().GetProperties())
{
var value1 = objParam.GetValue(firstObj);
var value2 = objParam.GetValue(secondObj);
if (value1 == null || value2 == null || !value1.Equals(value2))
{
logger.Error("Property: " + objParam.Name);
logger.Error("Values: " + value1?.ToString() + " and " + value2?.ToString());
result = false;
}
}
return result;
It works perfectly for the first level of params but it ignores completely any nested objects. In this example I would like it to compare the values inside the parameters object and if they are different the log to print error "Property: Parameters.Status".
I would recommend to look into some tool which already does that (do not know one which does exactly that but FluentAssertions for example can handle object graph comparisons). But in the nutshell you can check if type is primitive or overrides Equals and call your method recursively. Something like the following:
bool Compare(object firstObj, object secondObj)
{
if (object.ReferenceEquals(firstObj, secondObj))
{
return true;
}
var type = firstObj.GetType();
var propertyInfos = firstObj.GetType().GetProperties();
foreach (var objParam in propertyInfos)
{
var methodInfo = objParam.PropertyType.GetMethod(nameof(object.Equals), BindingFlags.Public | BindingFlags.Instance | BindingFlags.DeclaredOnly, new []{typeof(object)});
var overridesEquals = methodInfo?.GetBaseDefinition().DeclaringType == typeof(object);
var value1 = objParam.GetValue(firstObj);
var value2 = objParam.GetValue(secondObj);
if (value1 == null || value2 == null)
{
// Log
return false;
}
if (object.ReferenceEquals(value1, value2))
{
continue;
}
if (type.IsPrimitive || overridesEquals)
{
if (value1.Equals(value2))
{
continue;
}
// Log?
return false;
}
if (!Compare(value1, value2))
{
// log ?
return false;
}
}
return true;
}
P.S.
Note that Connection.ParametersObj is not a property it is field so it will be ignored by both yours and mine implementations
Consider using source generators instead of reflection.
This does not handle collections.
The problem is that you only do one loop, without looking at the objects within each object. Here's a quick recursive function I threw together. (untested!)
// You can make it non-generic, this just ensures that both arguments are the same type.
static void Go<T>(T left, T right, Action<object, object, PropertyInfo> onFound, int depth = 2)
{
if (left is null)
return;
foreach (var p in left.GetType().GetProperties())
{
var l = p.GetValue(left);
var r = p.GetValue(right);
if (l is null || r is null || !l.Equals(r))
onFound(l, r, p);
if (depth is not 0)
Go(l, r, onFound, depth - 1);
}
}
Usage:
var arg1 = new Connection()
{
ParametersObj = new() { AssignedName = "foo" }
};
var arg2 = new Connection()
{
ParametersObj = new() { AssignedName = "bar" }
};
Go(arg1, arg2, Log); // goes 2 layers deep is you don't specify the last parameter
void Log(object l, object r, PropertyInfo p)
{
logger.Error($"Property: {p.Name}");
logger.Error($"Values: {l} and {r}");
}

How to determine if property is a nullable reference type? [duplicate]

C# 8.0 introduces nullable reference types. Here's a simple class with a nullable property:
public class Foo
{
public String? Bar { get; set; }
}
Is there a way to check a class property uses a nullable reference type via reflection?
In .NET 6, the NullabilityInfoContext APIs were added to handle this. See this answer.
Prior to this, you need to read the attributes yourself. This appears to work, at least on the types I've tested it with.
public static bool IsNullable(PropertyInfo property) =>
IsNullableHelper(property.PropertyType, property.DeclaringType, property.CustomAttributes);
public static bool IsNullable(FieldInfo field) =>
IsNullableHelper(field.FieldType, field.DeclaringType, field.CustomAttributes);
public static bool IsNullable(ParameterInfo parameter) =>
IsNullableHelper(parameter.ParameterType, parameter.Member, parameter.CustomAttributes);
private static bool IsNullableHelper(Type memberType, MemberInfo? declaringType, IEnumerable<CustomAttributeData> customAttributes)
{
if (memberType.IsValueType)
return Nullable.GetUnderlyingType(memberType) != null;
var nullable = customAttributes
.FirstOrDefault(x => x.AttributeType.FullName == "System.Runtime.CompilerServices.NullableAttribute");
if (nullable != null && nullable.ConstructorArguments.Count == 1)
{
var attributeArgument = nullable.ConstructorArguments[0];
if (attributeArgument.ArgumentType == typeof(byte[]))
{
var args = (ReadOnlyCollection<CustomAttributeTypedArgument>)attributeArgument.Value!;
if (args.Count > 0 && args[0].ArgumentType == typeof(byte))
{
return (byte)args[0].Value! == 2;
}
}
else if (attributeArgument.ArgumentType == typeof(byte))
{
return (byte)attributeArgument.Value! == 2;
}
}
for (var type = declaringType; type != null; type = type.DeclaringType)
{
var context = type.CustomAttributes
.FirstOrDefault(x => x.AttributeType.FullName == "System.Runtime.CompilerServices.NullableContextAttribute");
if (context != null &&
context.ConstructorArguments.Count == 1 &&
context.ConstructorArguments[0].ArgumentType == typeof(byte))
{
return (byte)context.ConstructorArguments[0].Value! == 2;
}
}
// Couldn't find a suitable attribute
return false;
}
See this document for details.
The general gist is that either the property itself can have a [Nullable] attribute on it, or if it doesn't the enclosing type might have [NullableContext] attribute. We first look for [Nullable], then if we don't find it we look for [NullableContext] on the enclosing type.
The compiler might embed the attributes into the assembly, and since we might be looking at a type from a different assembly, we need to do a reflection-only load.
[Nullable] might be instantiated with an array, if the property is generic. In this case, the first element represents the actual property (and further elements represent generic arguments). [NullableContext] is always instantiated with a single byte.
A value of 2 means "nullable". 1 means "not nullable", and 0 means "oblivious".
.NET 6 Preview 7 adds reflection APIs to get nullability info.
Libraries: Reflection APIs for nullability information
Obviously, this only helps folks targeting .NET 6+.
Getting top-level nullability information
Imagine you’re implementing a serializer. Using these new APIs the serializer can check whether a given property can be set to null:
private NullabilityInfoContext _nullabilityContext = new NullabilityInfoContext();
private void DeserializePropertyValue(PropertyInfo p, object instance, object? value)
{
if (value is null)
{
var nullabilityInfo = _nullabilityContext.Create(p);
if (nullabilityInfo.WriteState is not NullabilityState.Nullable)
{
throw new MySerializerException($"Property '{p.GetType().Name}.{p.Name}'' cannot be set to null.");
}
}
p.SetValue(instance, value);
}
I wrote a library to do reflection of NRT types - internally it looks at the generated attributes and gives you a simple API:
https://github.com/RicoSuter/Namotion.Reflection
Late answer.
This is what I ended up using thanks to Bill Menees:
static bool IsMarkedAsNullable(PropertyInfo p)
{
return new NullabilityInfoContext().Create(p).WriteState is NullabilityState.Nullable;
}
// Tests:
class Foo
{
public int Int1 { get; set; }
public int? Int2 { get; set; } = null;
public string Str1 { get; set; } = "";
public string? Str2 { get; set; } = null;
public List<Foo> Lst1 { get; set; } = new();
public List<Foo>? Lst2 { get; set; } = null;
public Dictionary<int, object> Dic1 { get; set; } = new();
public Dictionary<int, object>? Dic2 { get; set; } = null;
}
....
var props = typeof(Foo).GetProperties();
foreach(var prop in props)
{
Console.WriteLine($"Prop:{prop.Name} IsNullable:{IsMarkedAsNullable(prop)}");
}
// outputs:
Prop:Int1 IsNullable:False
Prop:Int2 IsNullable:True
Prop:Str1 IsNullable:False
Prop:Str2 IsNullable:True
Prop:Lst1 IsNullable:False
Prop:Lst2 IsNullable:True
Prop:Dic1 IsNullable:False
Prop:Dic2 IsNullable:True
A great answer there by #rico-suter !
The following is for those who just want a quick cut-and-paste solution until the real McCoy is available (see the proposal https://github.com/dotnet/runtime/issues/29723 ).
I put this together based on #canton7's post above plus a short look at the ideas in #rico-suter's code. The change from the #canton7's code is just abstracting the list of attribute sources and including a few new ones.
private static bool IsAttributedAsNonNullable(this PropertyInfo propertyInfo)
{
return IsAttributedAsNonNullable(
new dynamic?[] { propertyInfo },
new dynamic?[] { propertyInfo.DeclaringType, propertyInfo.DeclaringType?.DeclaringType, propertyInfo.DeclaringType?.GetTypeInfo() }
);
}
private static bool IsAttributedAsNonNullable(this ParameterInfo parameterInfo)
{
return IsAttributedAsNonNullable(
new dynamic?[] { parameterInfo },
new dynamic?[] { parameterInfo.Member, parameterInfo.Member.DeclaringType, parameterInfo.Member.DeclaringType?.DeclaringType, parameterInfo.Member.DeclaringType?.GetTypeInfo()
);
}
private static bool IsAttributedAsNonNullable( dynamic?[] nullableAttributeSources, dynamic?[] nullableContextAttributeSources)
{
foreach (dynamic? nullableAttributeSource in nullableAttributeSources) {
if (nullableAttributeSource == null) { continue; }
CustomAttributeData? nullableAttribute = ((IEnumerable<CustomAttributeData>)nullableAttributeSource.CustomAttributes).FirstOrDefault(x => x.AttributeType.FullName == "System.Runtime.CompilerServices.NullableAttribute");
if (nullableAttribute != null && nullableAttribute.ConstructorArguments.Count == 1) {
CustomAttributeTypedArgument attributeArgument = nullableAttribute.ConstructorArguments[0];
if (attributeArgument.ArgumentType == typeof(byte[])) {
var args = (ReadOnlyCollection<CustomAttributeTypedArgument>)(attributeArgument.Value ?? throw new NullReferenceException("Logic error!"));
if (args.Count > 0 && args[0].ArgumentType == typeof(byte)) {
byte value = (byte)(args[0].Value ?? throw new NullabilityLogicException());
return value == 1; // 0 = oblivious, 1 = nonnullable, 2 = nullable
}
} else if (attributeArgument.ArgumentType == typeof(byte)) {
byte value = (byte)(attributeArgument.Value ?? throw new NullReferenceException("Logic error!"));
return value == 1; // 0 = oblivious, 1 = nonnullable, 2 = nullable
} else {
throw new InvalidOperationException($"Unrecognized argument type for NullableAttribute.");
}
}
}
foreach (dynamic? nullableContextAttributeSource in nullableContextAttributeSources) {
if (nullableContextAttributeSource == null) { continue; }
CustomAttributeData? nullableContextAttribute = ((IEnumerable<CustomAttributeData>)nullableContextAttributeSource.CustomAttributes).FirstOrDefault(x => x.AttributeType.FullName == "System.Runtime.CompilerServices.NullableContextAttribute");
if (nullableContextAttribute != null && nullableContextAttribute.ConstructorArguments.Count == 1) {
CustomAttributeTypedArgument attributeArgument = nullableContextAttribute.ConstructorArguments[0];
if (attributeArgument.ArgumentType == typeof(byte)) {
byte value = (byte)(nullableContextAttribute.ConstructorArguments[0].Value ?? throw new NullabilityLogicException());
return value == 1;
} else {
throw new InvalidOperationException($"Unrecognized argument type for NullableContextAttribute.");
}
}
}
return false;
}
It's only the string? which gets a bit tricky. The rest of the nullable types are pretty straightforward to find out. For strings I used the following method, which you need to pass in a PropertyInfo object taken via reflection.
private bool IsNullable(PropertyInfo prop)
{
return prop.CustomAttributes.Any(x => x.AttributeType.Name == "NullableAttribute");
}

Howto search through Properties of all kinds of types

I have a base class called Part and derived classes like Wire or Connector and many more that inherit from Part.
Now I want to implement a search function that searches all Properties of the derived classes for a string.
If necessary that string should be tried to be converted to the type of the Property. The Properties can also be Lists and should be searched on the first level.
class Part
{
public int Id { get; set; }
public string Name { get; set; }
}
class Wire : Part
{
public NumberWithUnit Diameter { get; set; }
public Weight Weight { get; set; }
}
class Connector : Part
{
public List<Part> ConnectedParts { get; set; }
}
I know how to generally search through the Properties of base types with Reflection like this
private bool SearchProperties<T>(T part, string searchString) where T : Part
{
var props = typeof(T).GetProperties();
foreach (var prop in props)
{
var value = prop.GetValue(part);
if (value is string)
{
if (string.Equals(value, searchString))
return true;
}
else if (value is int)
{
int v;
if (int.TryParse(searchString, out v))
{
if(v == (int) value)
return true;
}
}
}
return false;
}
But that would be a long list of types and I have Properties of Type Weight for instance and many more. Is there some kind of general way to search without casting all types?
Consider going the opposite direction with your conversion. Rather than converting your search string into each possible value, just convert the value into a string:
private bool SearchProperties<T>(T part, string searchString) where T : Part
{
var props = typeof(T).GetProperties();
foreach (var prop in props)
{
var value = prop.GetValue(part);
if (value is IEnumerable)
{
// special handling for collections
}
else if(value != null)
{
string valueString = value.ToString();
if (string.Equals(valueString, searchString))
return true;
}
}
return false;
}
Besides working pretty well for most built-in types, the only thing you have to do to get it to work for Weight, etc. is make sure they implement ToString().
Another solution would be to use TypeDescriptor:
private bool SearchProperties<T>(T part, string searchString) where T : Part
{
var props = typeof(T).GetProperties();
foreach (var prop in props)
{
var value = prop.GetValue(part);
if (value is IEnumerable)
{
// special handling for collections
}
else if(value != null)
{
object searchValue = null;
try
{
searchValue = TypeDescriptor.GetConverter(value).ConvertFromString(searchString);
} catch {}
if (searchValue != null && object.Equals(value, searchValue))
return true;
}
}
return false;
}
TypeDescriptor works well for most built-in types, but requires extra work if you're dealing with custom types.
I think the following should cover the most of the practical scenarios:
public static bool SearchProperties(object target, string searchString)
{
if (target == null) return false;
// Common types
var convertible = target as IConvertible;
if (convertible != null)
{
var typeCode = convertible.GetTypeCode();
if (typeCode == TypeCode.String) return target.ToString() == searchString;
if (typeCode == TypeCode.DBNull) return false;
if (typeCode != TypeCode.Object)
{
try
{
var value = Convert.ChangeType(searchString, typeCode);
return target.Equals(value);
}
catch { return false; }
}
}
if (target is DateTimeOffset)
{
DateTimeOffset value;
return DateTimeOffset.TryParse(searchString, out value) && value == (DateTimeOffset)target;
}
var enumerable = target as IEnumerable;
if (enumerable != null)
{
// Collection
foreach (var item in enumerable)
if (SearchProperties(item, searchString)) return true;
}
else
{
// Complex type
var properties = target.GetType().GetProperties();
foreach (var property in properties)
{
if (property.GetMethod == null || property.GetMethod.GetParameters().Length > 0) continue;
var value = property.GetValue(target);
if (SearchProperties(value, searchString)) return true;
}
}
return false;
}
I will give you one different idea to do it.
You could try something like that:
private bool SearchProperties<T, W>(T part, W searchValue) where T : Part
{
var props = typeof(T).GetProperties();
foreach (var prop in props)
{
if (typeof(W) == prop.PropertyType)
{
var value = prop.GetValue(part, null);
if (searchValue.Equals(value))
return true;
}
}
return false;
}
You need to call the method like this:
private void button12_Click(object sender, EventArgs e)
{
Part p = new Part();
p.Id = 2;
p.Name = "test";
p.bla = new Bla();
SearchProperties<Part, int>(p, 2);
}
And if you need to compare the complex properties (Weight, ...) by a different way from GetHashCode you could override the method Equals or the == operator.
class Weight
{
public int Id { get; set; }
public override bool Equals(object obj)
{
return Id == ((Weight)obj).Id;
}
}

Arg<object>.Is.Equal with anonymous objects

In my MVC3 project, I use an IUrlProvider interface to wrap the UrlHelper class. In one of my controller actions, I have a call like this:
string url = _urlProvider.Action("ValidateCode", new { code = "spam-and-eggs" });
I want to stub this method call in my unit test, which is in a separate project. The test setup looks something like this:
IUrlProvider urlProvider = MockRepository.GenerateStub<IUrlProvider>();
urlProvider.Stub(u => u.Action(
Arg<string>.Is.Equal("ValidateCode"),
Arg<object>.Is.Equal(new { code = "spam-and-eggs" }) ))
.Return("http://www.mysite.com/validate/spam-and-eggs");
Unfortunately, Arg<object>.Is.Equal(new { code = "spam-and-eggs" }) doesn't work, because new { code = "spam-and-eggs" } != new { code = "spam-and-eggs" } when the anonymous types are declared in different assemblies.
So, is there an alternative syntax I can use with Rhino Mocks to check for matching field values between anonymous objects across assemblies?
Or should I replace the anonymous object declarations with a class, like this?
public class CodeArg
{
public string code { get; set; }
public override bool Equals(object obj)
{
if(obj == null || GetType() != obj.GetType())
{
return false;
}
return code == ((CodeArg)obj).code;
}
public override int GetHashCode()
{
return code.GetHashCode();
}
}
string url = _urlProvider.Action("ValidateCode",
new CodeArg { code = "spam-and-eggs" });
IUrlProvider urlProvider = MockRepository.GenerateStub<IUrlProvider>();
urlProvider.Stub(u => u.Action(
Arg<string>.Is.Equal("ValidateCode"),
Arg<CodeArg>.Is.Equal(new CodeArg { code = "spam-and-eggs" }) ))
.Return("http://www.mysite.com/validate/spam-and-eggs");
EDIT
If my unit test was in the same project as my controller, comparing the anonymous objects would work fine. Because they are declared in separate assemblies, they will not be equal, even if they have the same field names and values. Comparing anonymous objects created by methods in different namespaces doesn't seem to be a problem.
SOLUTION
I replaced Arg<object>.Is.Equal() with Arg<object>.Matches() using a custom AbstractConstraint:
IUrlProvider urlProvider = MockRepository.GenerateStub<IUrlProvider>();
urlProvider.Stub(u => u.Action(
Arg<string>.Is.Equal("ValidateCode"),
Arg<object>.Matches(new PropertiesMatchConstraint(new { code = "spam-and-eggs" })) ))
.Return("http://www.mysite.com/validate/spam-and-eggs");
public class PropertiesMatchConstraint : AbstractConstraint
{
private readonly object _equal;
public PropertiesMatchConstraint(object obj)
{
_equal = obj;
}
public override bool Eval(object obj)
{
if (obj == null)
{
return (_equal == null);
}
var equalType = _equal.GetType();
var objType = obj.GetType();
foreach (var property in equalType.GetProperties())
{
var otherProperty = objType.GetProperty(property.Name);
if (otherProperty == null || property.GetValue(_equal, null) != otherProperty.GetValue(obj, null))
{
return false;
}
}
return true;
}
public override string Message
{
get
{
string str = _equal == null ? "null" : _equal.ToString();
return "equal to " + str;
}
}
}
Anonymous types do implement Equals and GetHashCode in a pretty normal way, calling GetHashCode and Equals for each of their submembers.
So this should pass:
Assert.AreEqual(new { code = "spam-and-eggs" },
new { code = "spam-and-eggs" });
In other words, I suspect you're looking for the problem in the wrong place.
Note that you have to specify the properties in exactly the right order - so new { a = 0, b = 1 } will not be equal to new { b = 1, a = 0 }; the two objects will be of different types.
EDIT: The anonymous type instance creation expressions have to be in the same assembly, too. This is no doubt the problem in this case.
If Equals allows you to specify an IEqualityComparer<T>, you could probably build one which is able to compare two anonymous types with the same properties by creating an instance of one type from the properties of an instance of the other, and then comparing that to the original of the same type. Of course if you were using nested anonymous types you'd need to do that recursively, which could get ugly...
As GetValue returns a boxed value, this appears to work correctly.
public class PropertiesMatchConstraint : AbstractConstraint
{
private readonly object _equal;
public PropertiesMatchConstraint(object obj)
{
_equal = obj;
}
public override bool Eval(object obj)
{
if (obj == null)
{
return (_equal == null);
}
var equalType = _equal.GetType();
var objType = obj.GetType();
foreach (var property in equalType.GetProperties())
{
var otherProperty = objType.GetProperty(property.Name);
if (otherProperty == null || !_ValuesMatch(property.GetValue(_equal, null), otherProperty.GetValue(obj, null)))
{
return false;
}
}
return true;
}
//fix for boxed conversions object:Guid != object:Guid when both values are the same guid - must call .Equals()
private bool _ValuesMatch(object value, object otherValue)
{
if (value == otherValue)
return true; //return early
if (value != null)
return value.Equals(otherValue);
return otherValue.Equals(value);
}
public override string Message
{
get
{
string str = _equal == null ? "null" : _equal.ToString();
return "equal to " + str;
}
}
}

Using reflection in C# to get properties of a nested object

Given the following objects:
public class Customer {
public String Name { get; set; }
public String Address { get; set; }
}
public class Invoice {
public String ID { get; set; }
public DateTime Date { get; set; }
public Customer BillTo { get; set; }
}
I'd like to use reflection to go through the Invoice to get the Name property of a Customer. Here's what I'm after, assuming this code would work:
Invoice inv = GetDesiredInvoice(); // magic method to get an invoice
PropertyInfo info = inv.GetType().GetProperty("BillTo.Address");
Object val = info.GetValue(inv, null);
Of course, this fails since "BillTo.Address" is not a valid property of the Invoice class.
So, I tried writing a method to split the string into pieces on the period, and walk the objects looking for the final value I was interested in. It works okay, but I'm not entirely comfortable with it:
public Object GetPropValue(String name, Object obj) {
foreach (String part in name.Split('.')) {
if (obj == null) { return null; }
Type type = obj.GetType();
PropertyInfo info = type.GetProperty(part);
if (info == null) { return null; }
obj = info.GetValue(obj, null);
}
return obj;
}
Any ideas on how to improve this method, or a better way to solve this problem?
EDIT after posting, I saw a few related posts... There doesn't seem to be an answer that specifically addresses this question, however. Also, I'd still like the feedback on my implementation.
I use following method to get the values from (nested classes) properties like
"Property"
"Address.Street"
"Address.Country.Name"
public static object GetPropertyValue(object src, string propName)
{
if (src == null) throw new ArgumentException("Value cannot be null.", "src");
if (propName == null) throw new ArgumentException("Value cannot be null.", "propName");
if(propName.Contains("."))//complex type nested
{
var temp = propName.Split(new char[] { '.' }, 2);
return GetPropertyValue(GetPropertyValue(src, temp[0]), temp[1]);
}
else
{
var prop = src.GetType().GetProperty(propName);
return prop != null ? prop.GetValue(src, null) : null;
}
}
Here is the Fiddle: https://dotnetfiddle.net/PvKRH0
I know I'm a bit late to the party, and as others said, your implementation is fine
...for simple use cases.
However, I've developed a library that solves exactly that use case, Pather.CSharp.
It is also available as Nuget Package.
Its main class is Resolver with its Resolve method.
You pass it an object and the property path, and it will return the desired value.
Invoice inv = GetDesiredInvoice(); // magic method to get an invoice
var resolver = new Resolver();
object result = resolver.Resolve(inv, "BillTo.Address");
But it can also resolve more complex property paths, including array and dictionary access.
So, for example, if your Customer had multiple addresses
public class Customer {
public String Name { get; set; }
public IEnumerable<String> Addresses { get; set; }
}
you could access the second one using Addresses[1].
Invoice inv = GetDesiredInvoice(); // magic method to get an invoice
var resolver = new Resolver();
object result = resolver.Resolve(inv, "BillTo.Addresses[1]");
I actually think your logic is fine. Personally, I would probably change it around so you pass the object as the first parameter (which is more inline with PropertyInfo.GetValue, so less surprising).
I also would probably call it something more like GetNestedPropertyValue, to make it obvious that it searches down the property stack.
You have to access the ACTUAL object that you need to use reflection on. Here is what I mean:
Instead of this:
Invoice inv = GetDesiredInvoice(); // magic method to get an invoice
PropertyInfo info = inv.GetType().GetProperty("BillTo.Address");
Object val = info.GetValue(inv, null);
Do this (edited based on comment):
Invoice inv = GetDesiredInvoice(); // magic method to get an invoice
PropertyInfo info = inv.GetType().GetProperty("BillTo");
Customer cust = (Customer)info.GetValue(inv, null);
PropertyInfo info2 = cust.GetType().GetProperty("Address");
Object val = info2.GetValue(cust, null);
Look at this post for more information:
Using reflection to set a property of a property of an object
In hopes of not sounding too late to the party, I would like to add my solution:
Definitely use recursion in this situation
public static Object GetPropValue(String name, object obj, Type type)
{
var parts = name.Split('.').ToList();
var currentPart = parts[0];
PropertyInfo info = type.GetProperty(currentPart);
if (info == null) { return null; }
if (name.IndexOf(".") > -1)
{
parts.Remove(currentPart);
return GetPropValue(String.Join(".", parts), info.GetValue(obj, null), info.PropertyType);
} else
{
return info.GetValue(obj, null).ToString();
}
}
You don't explain the source of your "discomfort," but your code basically looks sound to me.
The only thing I'd question is the error handling. You return null if the code tries to traverse through a null reference or if the property name doesn't exist. This hides errors: it's hard to know whether it returned null because there's no BillTo customer, or because you misspelled it "BilTo.Address"... or because there is a BillTo customer, and its Address is null! I'd let the method crash and burn in these cases -- just let the exception escape (or maybe wrap it in a friendlier one).
Here is another implementation that will skip a nested property if it is an enumerator and continue deeper. Properties of type string are not affected by the Enumeration Check.
public static class ReflectionMethods
{
public static bool IsNonStringEnumerable(this PropertyInfo pi)
{
return pi != null && pi.PropertyType.IsNonStringEnumerable();
}
public static bool IsNonStringEnumerable(this object instance)
{
return instance != null && instance.GetType().IsNonStringEnumerable();
}
public static bool IsNonStringEnumerable(this Type type)
{
if (type == null || type == typeof(string))
return false;
return typeof(IEnumerable).IsAssignableFrom(type);
}
public static Object GetPropValue(String name, Object obj)
{
foreach (String part in name.Split('.'))
{
if (obj == null) { return null; }
if (obj.IsNonStringEnumerable())
{
var toEnumerable = (IEnumerable)obj;
var iterator = toEnumerable.GetEnumerator();
if (!iterator.MoveNext())
{
return null;
}
obj = iterator.Current;
}
Type type = obj.GetType();
PropertyInfo info = type.GetProperty(part);
if (info == null) { return null; }
obj = info.GetValue(obj, null);
}
return obj;
}
}
based on this question and on
How to know if a PropertyInfo is a collection
by Berryl
I use this in a MVC project to dynamically Order my data by simply passing the Property to sort by
Example:
result = result.OrderBy((s) =>
{
return ReflectionMethods.GetPropValue("BookingItems.EventId", s);
}).ToList();
where BookingItems is a list of objects.
> Get Nest properties e.g., Developer.Project.Name
private static System.Reflection.PropertyInfo GetProperty(object t, string PropertName)
{
if (t.GetType().GetProperties().Count(p => p.Name == PropertName.Split('.')[0]) == 0)
throw new ArgumentNullException(string.Format("Property {0}, is not exists in object {1}", PropertName, t.ToString()));
if (PropertName.Split('.').Length == 1)
return t.GetType().GetProperty(PropertName);
else
return GetProperty(t.GetType().GetProperty(PropertName.Split('.')[0]).GetValue(t, null), PropertName.Split('.')[1]);
}
if (info == null) { /* throw exception instead*/ }
I would actually throw an exception if they request a property that doesn't exist. The way you have it coded, if I call GetPropValue and it returns null, I don't know if that means the property didn't exist, or the property did exist but it's value was null.
public static string GetObjectPropertyValue(object obj, string propertyName)
{
bool propertyHasDot = propertyName.IndexOf(".") > -1;
string firstPartBeforeDot;
string nextParts = "";
if (!propertyHasDot)
firstPartBeforeDot = propertyName.ToLower();
else
{
firstPartBeforeDot = propertyName.Substring(0, propertyName.IndexOf(".")).ToLower();
nextParts = propertyName.Substring(propertyName.IndexOf(".") + 1);
}
foreach (var property in obj.GetType().GetProperties())
if (property.Name.ToLower() == firstPartBeforeDot)
if (!propertyHasDot)
if (property.GetValue(obj, null) != null)
return property.GetValue(obj, null).ToString();
else
return DefaultValue(property.GetValue(obj, null), propertyName).ToString();
else
return GetObjectPropertyValue(property.GetValue(obj, null), nextParts);
throw new Exception("Property '" + propertyName.ToString() + "' not found in object '" + obj.ToString() + "'");
}
I wanted to share my solution although it may be too late. This solution is primarily to check if the nested property exists. But it can be easily tweaked to return the property value if needed.
private static PropertyInfo _GetPropertyInfo(Type type, string propertyName)
{
//***
//*** Check if the property name is a complex nested type
//***
if (propertyName.Contains("."))
{
//***
//*** Get the first property name of the complex type
//***
var tempPropertyName = propertyName.Split(".", 2);
//***
//*** Check if the property exists in the type
//***
var prop = _GetPropertyInfo(type, tempPropertyName[0]);
if (prop != null)
{
//***
//*** Drill down to check if the nested property exists in the complex type
//***
return _GetPropertyInfo(prop.PropertyType, tempPropertyName[1]);
}
else
{
return null;
}
}
else
{
return type.GetProperty(propertyName, BindingFlags.IgnoreCase | BindingFlags.Public | BindingFlags.Instance);
}
}
I had to refer to few posts to come up with this solution. I think this will work for multiple nested property types.
My internet connection was down when I need to solve the same problem, so I had to 're-invent the wheel':
static object GetPropertyValue(Object fromObject, string propertyName)
{
Type objectType = fromObject.GetType();
PropertyInfo propInfo = objectType.GetProperty(propertyName);
if (propInfo == null && propertyName.Contains('.'))
{
string firstProp = propertyName.Substring(0, propertyName.IndexOf('.'));
propInfo = objectType.GetProperty(firstProp);
if (propInfo == null)//property name is invalid
{
throw new ArgumentException(String.Format("Property {0} is not a valid property of {1}.", firstProp, fromObject.GetType().ToString()));
}
return GetPropertyValue(propInfo.GetValue(fromObject, null), propertyName.Substring(propertyName.IndexOf('.') + 1));
}
else
{
return propInfo.GetValue(fromObject, null);
}
}
Pretty sure this solves the problem for any string you use for property name, regardless of extent of nesting, as long as everything's a property.
Based on the original code from #jheddings, I have created a extension method version with generic type and verifications:
public static T GetPropertyValue<T>(this object sourceObject, string propertyName)
{
if (sourceObject == null) throw new ArgumentNullException(nameof(sourceObject));
if (string.IsNullOrWhiteSpace(propertyName)) throw new ArgumentException(nameof(propertyName));
foreach (string currentPropertyName in propertyName.Split('.'))
{
if (string.IsNullOrWhiteSpace(currentPropertyName)) throw new InvalidOperationException($"Invalid property '{propertyName}'");
PropertyInfo propertyInfo = sourceObject.GetType().GetProperty(currentPropertyName);
if (propertyInfo == null) throw new InvalidOperationException($"Property '{currentPropertyName}' not found");
sourceObject = propertyInfo.GetValue(sourceObject);
}
return sourceObject is T result ? result : default;
}
I wrote a method that received one object type as the argument from the input and returns dictionary<string,string>
public static Dictionary<string, string> GetProperties(Type placeHolderType)
{
var result = new Dictionary<string, string>();
var properties = placeHolderType.GetProperties();
foreach (var propertyInfo in properties)
{
string name = propertyInfo.Name;
string description = GetDescriptionTitle(propertyInfo);
if (IsNonString(propertyInfo.PropertyType))
{
var list = GetProperties(propertyInfo.PropertyType);
foreach (var item in list)
{
result.Add($"{propertyInfo.PropertyType.Name}_{item.Key}", item.Value);
}
}
else
{
result.Add(name, description);
}
}
return result;
}
public static bool IsNonString(Type type)
{
if (type == null || type == typeof(string))
return false;
return typeof(IPlaceHolder).IsAssignableFrom(type);
}
private static string GetDescriptionTitle(MemberInfo memberInfo)
{
if (Attribute.GetCustomAttribute(memberInfo, typeof(DescriptionAttribute)) is DescriptionAttribute descriptionAttribute)
{
return descriptionAttribute.Description;
}
return memberInfo.Name;
}
public static object GetPropertyValue(object src, string propName)
{
if (src == null) throw new ArgumentException("Value cannot be null.", "src");
if (propName == null) throw new ArgumentException("Value cannot be null.", "propName");
var prop = src.GetType().GetProperty(propName);
if (prop != null)
{
return prop.GetValue(src, null);
}
else
{
var props = src.GetType().GetProperties();
foreach (var property in props)
{
var propInfo = src.GetType().GetProperty(property.Name);
if (propInfo != null)
{
var propVal = propInfo.GetValue(src, null);
if (src.GetType().GetProperty(property.Name).PropertyType.IsClass)
{
return GetPropertyValue(propVal, propName);
}
return propVal;
}
}
return null;
}
usage: calling part
var emp = new Employee() { Person = new Person() { FirstName = "Ashwani" } };
var val = GetPropertyValue(emp, "FirstName");
above can search the property value at any level
Try inv.GetType().GetProperty("BillTo+Address");

Categories