Member variable becomes null, spuriously? - grails

I observe an assertion failure which I've attempted to sketch below. Briefly, an initialized class member becomes null spuriously when I call an overridden method.
abstract class A<T> {
T target
def something() {
assert target != null // sanity check -- fails when invoked below
return this
}
}
class B extends A<Something> {
static B makeB() {
def b = new B()
b.target = new Something();
assert b.target != null
return b
}
}
// package test
// Test code
//
def b = B.makeB() // <-- success.
assert b != null
assert b.target != null
b.something() // <-- fails assertion
print "Success!"
I'm having trouble producing a minimal test case -- the above (with Long in place of Something) runs successfully at compileonline.com but not with our code on our machines.
The problem goes away if I re-implement the something method in the subclass B.
Has anybody observed a similar effect? I'm already quite frustrated at this.

Problem Found: A Mixin uses the same member name
Stepping through the code and examining local stack variables carefully, we found that the Grails Integration Test Mixin has an instance variable target that was receiving the assignment messages. Presumably, any instance variable sharing a name with a mixin's instance variable should be subject to the behavior I describe above, but we don't know the complete set of conditions
The most reasonably solution time-wise (with a nod #cfrick) was to rename our variable target to something else.
Follow-up Questions
Is this expected behavior? Best solution in a sea of bad ones?
Has this behavior changed with a new Groovy compiler or Grails stack?
(still relevant) What is the minimal test case?

Related

An exception of type 'System.NullReferenceException' occurred in Proj.Web.dll but was not handled in user code [duplicate]

This question's answers are a community effort. Edit existing answers to improve this post. It is not currently accepting new answers or interactions.
I have some code and when it executes, it throws a NullReferenceException, saying:
Object reference not set to an instance of an object.
What does this mean, and what can I do to fix this error?
What is the cause?
Bottom Line
You are trying to use something that is null (or Nothing in VB.NET). This means you either set it to null, or you never set it to anything at all.
Like anything else, null gets passed around. If it is null in method "A", it could be that method "B" passed a null to method "A".
null can have different meanings:
Object variables that are uninitialized and hence point to nothing. In this case, if you access members of such objects, it causes a NullReferenceException.
The developer is using null intentionally to indicate there is no meaningful value available. Note that C# has the concept of nullable datatypes for variables (like database tables can have nullable fields) - you can assign null to them to indicate there is no value stored in it, for example int? a = null; (which is a shortcut for Nullable<int> a = null;) where the question mark indicates it is allowed to store null in variable a. You can check that either with if (a.HasValue) {...} or with if (a==null) {...}. Nullable variables, like a this example, allow to access the value via a.Value explicitly, or just as normal via a. Note that accessing it via a.Value throws an InvalidOperationException instead of a NullReferenceException if a is null - you should do the check beforehand, i.e. if you have another non-nullable variable int b; then you should do assignments like if (a.HasValue) { b = a.Value; } or shorter if (a != null) { b = a; }.
The rest of this article goes into more detail and shows mistakes that many programmers often make which can lead to a NullReferenceException.
More Specifically
The runtime throwing a NullReferenceException always means the same thing: you are trying to use a reference, and the reference is not initialized (or it was once initialized, but is no longer initialized).
This means the reference is null, and you cannot access members (such as methods) through a null reference. The simplest case:
string foo = null;
foo.ToUpper();
This will throw a NullReferenceException at the second line because you can't call the instance method ToUpper() on a string reference pointing to null.
Debugging
How do you find the source of a NullReferenceException? Apart from looking at the exception itself, which will be thrown exactly at the location where it occurs, the general rules of debugging in Visual Studio apply: place strategic breakpoints and inspect your variables, either by hovering the mouse over their names, opening a (Quick)Watch window or using the various debugging panels like Locals and Autos.
If you want to find out where the reference is or isn't set, right-click its name and select "Find All References". You can then place a breakpoint at every found location and run your program with the debugger attached. Every time the debugger breaks on such a breakpoint, you need to determine whether you expect the reference to be non-null, inspect the variable, and verify that it points to an instance when you expect it to.
By following the program flow this way, you can find the location where the instance should not be null, and why it isn't properly set.
Examples
Some common scenarios where the exception can be thrown:
Generic
ref1.ref2.ref3.member
If ref1 or ref2 or ref3 is null, then you'll get a NullReferenceException. If you want to solve the problem, then find out which one is null by rewriting the expression to its simpler equivalent:
var r1 = ref1;
var r2 = r1.ref2;
var r3 = r2.ref3;
r3.member
Specifically, in HttpContext.Current.User.Identity.Name, the HttpContext.Current could be null, or the User property could be null, or the Identity property could be null.
Indirect
public class Person
{
public int Age { get; set; }
}
public class Book
{
public Person Author { get; set; }
}
public class Example
{
public void Foo()
{
Book b1 = new Book();
int authorAge = b1.Author.Age; // You never initialized the Author property.
// there is no Person to get an Age from.
}
}
If you want to avoid the child (Person) null reference, you could initialize it in the parent (Book) object's constructor.
Nested Object Initializers
The same applies to nested object initializers:
Book b1 = new Book
{
Author = { Age = 45 }
};
This translates to:
Book b1 = new Book();
b1.Author.Age = 45;
While the new keyword is used, it only creates a new instance of Book, but not a new instance of Person, so the Author the property is still null.
Nested Collection Initializers
public class Person
{
public ICollection<Book> Books { get; set; }
}
public class Book
{
public string Title { get; set; }
}
The nested collection Initializers behave the same:
Person p1 = new Person
{
Books = {
new Book { Title = "Title1" },
new Book { Title = "Title2" },
}
};
This translates to:
Person p1 = new Person();
p1.Books.Add(new Book { Title = "Title1" });
p1.Books.Add(new Book { Title = "Title2" });
The new Person only creates an instance of Person, but the Books collection is still null. The collection Initializer syntax does not create a collection
for p1.Books, it only translates to the p1.Books.Add(...) statements.
Array
int[] numbers = null;
int n = numbers[0]; // numbers is null. There is no array to index.
Array Elements
Person[] people = new Person[5];
people[0].Age = 20 // people[0] is null. The array was allocated but not
// initialized. There is no Person to set the Age for.
Jagged Arrays
long[][] array = new long[1][];
array[0][0] = 3; // is null because only the first dimension is yet initialized.
// Use array[0] = new long[2]; first.
Collection/List/Dictionary
Dictionary<string, int> agesForNames = null;
int age = agesForNames["Bob"]; // agesForNames is null.
// There is no Dictionary to perform the lookup.
Range Variable (Indirect/Deferred)
public class Person
{
public string Name { get; set; }
}
var people = new List<Person>();
people.Add(null);
var names = from p in people select p.Name;
string firstName = names.First(); // Exception is thrown here, but actually occurs
// on the line above. "p" is null because the
// first element we added to the list is null.
Events (C#)
public class Demo
{
public event EventHandler StateChanged;
protected virtual void OnStateChanged(EventArgs e)
{
StateChanged(this, e); // Exception is thrown here
// if no event handlers have been attached
// to StateChanged event
}
}
(Note: The VB.NET compiler inserts null checks for event usage, so it's not necessary to check events for Nothing in VB.NET.)
Bad Naming Conventions:
If you named fields differently from locals, you might have realized that you never initialized the field.
public class Form1
{
private Customer customer;
private void Form1_Load(object sender, EventArgs e)
{
Customer customer = new Customer();
customer.Name = "John";
}
private void Button_Click(object sender, EventArgs e)
{
MessageBox.Show(customer.Name);
}
}
This can be solved by following the convention to prefix fields with an underscore:
private Customer _customer;
ASP.NET Page Life cycle:
public partial class Issues_Edit : System.Web.UI.Page
{
protected TestIssue myIssue;
protected void Page_Load(object sender, EventArgs e)
{
if (!IsPostBack)
{
// Only called on first load, not when button clicked
myIssue = new TestIssue();
}
}
protected void SaveButton_Click(object sender, EventArgs e)
{
myIssue.Entry = "NullReferenceException here!";
}
}
ASP.NET Session Values
// if the "FirstName" session value has not yet been set,
// then this line will throw a NullReferenceException
string firstName = Session["FirstName"].ToString();
ASP.NET MVC empty view models
If the exception occurs when referencing a property of #Model in an ASP.NET MVC View, you need to understand that the Model gets set in your action method, when you return a view. When you return an empty model (or model property) from your controller, the exception occurs when the views access it:
// Controller
public class Restaurant:Controller
{
public ActionResult Search()
{
return View(); // Forgot the provide a Model here.
}
}
// Razor view
#foreach (var restaurantSearch in Model.RestaurantSearch) // Throws.
{
}
<p>#Model.somePropertyName</p> <!-- Also throws -->
WPF Control Creation Order and Events
WPF controls are created during the call to InitializeComponent in the order they appear in the visual tree. A NullReferenceException will be raised in the case of early-created controls with event handlers, etc., that fire during InitializeComponent which reference late-created controls.
For example:
<Grid>
<!-- Combobox declared first -->
<ComboBox Name="comboBox1"
Margin="10"
SelectedIndex="0"
SelectionChanged="comboBox1_SelectionChanged">
<ComboBoxItem Content="Item 1" />
<ComboBoxItem Content="Item 2" />
<ComboBoxItem Content="Item 3" />
</ComboBox>
<!-- Label declared later -->
<Label Name="label1"
Content="Label"
Margin="10" />
</Grid>
Here comboBox1 is created before label1. If comboBox1_SelectionChanged attempts to reference `label1, it will not yet have been created.
private void comboBox1_SelectionChanged(object sender, SelectionChangedEventArgs e)
{
label1.Content = comboBox1.SelectedIndex.ToString(); // NullReferenceException here!!
}
Changing the order of the declarations in the XAML (i.e., listing label1 before comboBox1, ignoring issues of design philosophy) would at least resolve the NullReferenceException here.
Cast with as
var myThing = someObject as Thing;
This doesn't throw an InvalidCastException but returns a null when the cast fails (and when someObject is itself null). So be aware of that.
LINQ FirstOrDefault() and SingleOrDefault()
The plain versions First() and Single() throw exceptions when there is nothing. The "OrDefault" versions return null in that case. So be aware of that.
foreach
foreach throws when you try to iterate on a null collection. Usually caused by unexpected null result from methods that return collections.
List<int> list = null;
foreach(var v in list) { } // NullReferenceException here
More realistic example - select nodes from XML document. Will throw if nodes are not found but initial debugging shows that all properties valid:
foreach (var node in myData.MyXml.DocumentNode.SelectNodes("//Data"))
Ways to Avoid
Explicitly check for null and ignore null values.
If you expect the reference sometimes to be null, you can check for it being null before accessing instance members:
void PrintName(Person p)
{
if (p != null)
{
Console.WriteLine(p.Name);
}
}
Explicitly check for null and provide a default value.
Methods you call expecting an instance can return null, for example when the object being sought cannot be found. You can choose to return a default value when this is the case:
string GetCategory(Book b)
{
if (b == null)
return "Unknown";
return b.Category;
}
Explicitly check for null from method calls and throw a custom exception.
You can also throw a custom exception, only to catch it in the calling code:
string GetCategory(string bookTitle)
{
var book = library.FindBook(bookTitle); // This may return null
if (book == null)
throw new BookNotFoundException(bookTitle); // Your custom exception
return book.Category;
}
Use Debug.Assert if a value should never be null, to catch the problem earlier than the exception occurs.
When you know during development that a method could, but never should return null, you can use Debug.Assert() to break as soon as possible when it does occur:
string GetTitle(int knownBookID)
{
// You know this should never return null.
var book = library.GetBook(knownBookID);
// Exception will occur on the next line instead of at the end of this method.
Debug.Assert(book != null, "Library didn't return a book for known book ID.");
// Some other code
return book.Title; // Will never throw NullReferenceException in Debug mode.
}
Though this check will not end up in your release build, causing it to throw the NullReferenceException again when book == null at runtime in release mode.
Use GetValueOrDefault() for nullable value types to provide a default value when they are null.
DateTime? appointment = null;
Console.WriteLine(appointment.GetValueOrDefault(DateTime.Now));
// Will display the default value provided (DateTime.Now), because appointment is null.
appointment = new DateTime(2022, 10, 20);
Console.WriteLine(appointment.GetValueOrDefault(DateTime.Now));
// Will display the appointment date, not the default
Use the null coalescing operator: ?? [C#] or If() [VB].
The shorthand to providing a default value when a null is encountered:
IService CreateService(ILogger log, Int32? frobPowerLevel)
{
var serviceImpl = new MyService(log ?? NullLog.Instance);
// Note that the above "GetValueOrDefault()" can also be rewritten to use
// the coalesce operator:
serviceImpl.FrobPowerLevel = frobPowerLevel ?? 5;
}
Use the null condition operator: ?. or ?[x] for arrays (available in C# 6 and VB.NET 14):
This is also sometimes called the safe navigation or Elvis (after its shape) operator. If the expression on the left side of the operator is null, then the right side will not be evaluated, and null is returned instead. That means cases like this:
var title = person.Title.ToUpper();
If the person does not have a title, this will throw an exception because it is trying to call ToUpper on a property with a null value.
In C# 5 and below, this can be guarded with:
var title = person.Title == null ? null : person.Title.ToUpper();
Now the title variable will be null instead of throwing an exception. C# 6 introduces a shorter syntax for this:
var title = person.Title?.ToUpper();
This will result in the title variable being null, and the call to ToUpper is not made if person.Title is null.
Of course, you still have to check title for null or use the null condition operator together with the null coalescing operator (??) to supply a default value:
// regular null check
int titleLength = 0;
if (title != null)
titleLength = title.Length; // If title is null, this would throw NullReferenceException
// combining the `?` and the `??` operator
int titleLength = title?.Length ?? 0;
Likewise, for arrays you can use ?[i] as follows:
int[] myIntArray = null;
var i = 5;
int? elem = myIntArray?[i];
if (!elem.HasValue) Console.WriteLine("No value");
This will do the following: If myIntArray is null, the expression returns null and you can safely check it. If it contains an array, it will do the same as:
elem = myIntArray[i]; and returns the ith element.
Use null context (available in C# 8):
Introduced in C# 8, null contexts and nullable reference types perform static analysis on variables and provide a compiler warning if a value can be potentially null or have been set to null. The nullable reference types allow types to be explicitly allowed to be null.
The nullable annotation context and nullable warning context can be set for a project using the Nullable element in your csproj file. This element configures how the compiler interprets the nullability of types and what warnings are generated. Valid settings are:
enable: The nullable annotation context is enabled. The nullable warning context is enabled. Variables of a reference type, string, for example, are non-nullable. All nullability warnings are enabled.
disable: The nullable annotation context is disabled. The nullable warning context is disabled. Variables of a reference type are oblivious, just like earlier versions of C#. All nullability warnings are disabled.
safeonly: The nullable annotation context is enabled. The nullable warning context is safeonly. Variables of a reference type are non-nullable. All safety nullability warnings are enabled.
warnings: The nullable annotation context is disabled. The nullable warning context is enabled. Variables of a reference type are oblivious. All nullability warnings are enabled.
safeonlywarnings: The nullable annotation context is disabled. The nullable warning context is safeonly.
Variables of a reference type are oblivious. All safety nullability warnings are enabled.
A nullable reference type is noted using the same syntax as nullable value types: a ? is appended to the type of the variable.
Special techniques for debugging and fixing null derefs in iterators
C# supports "iterator blocks" (called "generators" in some other popular languages). NullReferenceException can be particularly tricky to debug in iterator blocks because of deferred execution:
public IEnumerable<Frob> GetFrobs(FrobFactory f, int count)
{
for (int i = 0; i < count; ++i)
yield return f.MakeFrob();
}
...
FrobFactory factory = whatever;
IEnumerable<Frobs> frobs = GetFrobs();
...
foreach(Frob frob in frobs) { ... }
If whatever results in null then MakeFrob will throw. Now, you might think that the right thing to do is this:
// DON'T DO THIS
public IEnumerable<Frob> GetFrobs(FrobFactory f, int count)
{
if (f == null)
throw new ArgumentNullException("f", "factory must not be null");
for (int i = 0; i < count; ++i)
yield return f.MakeFrob();
}
Why is this wrong? Because the iterator block does not actually run until the foreach! The call to GetFrobs simply returns an object which when iterated will run the iterator block.
By writing a null check like this you prevent the NullReferenceException, but you move the NullArgumentException to the point of the iteration, not to the point of the call, and that is very confusing to debug.
The correct fix is:
// DO THIS
public IEnumerable<Frob> GetFrobs(FrobFactory f, int count)
{
// No yields in a public method that throws!
if (f == null)
throw new ArgumentNullException("f", "factory must not be null");
return GetFrobsForReal(f, count);
}
private IEnumerable<Frob> GetFrobsForReal(FrobFactory f, int count)
{
// Yields in a private method
Debug.Assert(f != null);
for (int i = 0; i < count; ++i)
yield return f.MakeFrob();
}
That is, make a private helper method that has the iterator block logic and a public surface method that does the null check and returns the iterator. Now when GetFrobs is called, the null check happens immediately, and then GetFrobsForReal executes when the sequence is iterated.
If you examine the reference source for LINQ to Objects you will see that this technique is used throughout. It is slightly more clunky to write, but it makes debugging nullity errors much easier. Optimize your code for the convenience of the caller, not the convenience of the author.
A note on null dereferences in unsafe code
C# has an "unsafe" mode which is, as the name implies, extremely dangerous because the normal safety mechanisms which provide memory safety and type safety are not enforced. You should not be writing unsafe code unless you have a thorough and deep understanding of how memory works.
In unsafe mode, you should be aware of two important facts:
dereferencing a null pointer produces the same exception as dereferencing a null reference
dereferencing an invalid non-null pointer can produce that exception in some circumstances
To understand why that is, it helps to understand how .NET produces NullReferenceException in the first place. (These details apply to .NET running on Windows; other operating systems use similar mechanisms.)
Memory is virtualized in Windows; each process gets a virtual memory space of many "pages" of memory that are tracked by the operating system. Each page of memory has flags set on it that determine how it may be used: read from, written to, executed, and so on. The lowest page is marked as "produce an error if ever used in any way".
Both a null pointer and a null reference in C# are internally represented as the number zero, and so any attempt to dereference it into its corresponding memory storage causes the operating system to produce an error. The .NET runtime then detects this error and turns it into the NullReferenceException.
That's why dereferencing both a null pointer and a null reference produces the same exception.
What about the second point? Dereferencing any invalid pointer that falls in the lowest page of virtual memory causes the same operating system error, and thereby the same exception.
Why does this make sense? Well, suppose we have a struct containing two ints, and an unmanaged pointer equal to null. If we attempt to dereference the second int in the struct, the CLR will not attempt to access the storage at location zero; it will access the storage at location four. But logically this is a null dereference because we are getting to that address via the null.
If you are working with unsafe code and you get a NullReferenceException, just be aware that the offending pointer need not be null. It can be any location in the lowest page, and this exception will be produced.
NullReference Exception — Visual Basic
The NullReference Exception for Visual Basic is no different from the one in C#. After all, they are both reporting the same exception defined in the .NET Framework which they both use. Causes unique to Visual Basic are rare (perhaps only one).
This answer will use Visual Basic terms, syntax, and context. The examples used come from a large number of past Stack  Overflow questions. This is to maximize relevance by using the kinds of situations often seen in posts. A bit more explanation is also provided for those who might need it. An example similar to yours is very likely listed here.
Note:
This is concept-based: there is no code for you to paste into your project. It is intended to help you understand what causes a NullReferenceException (NRE), how to find it, how to fix it, and how to avoid it. An NRE can be caused many ways so this is unlikely to be your sole encounter.
The examples (from Stack  Overflow posts) do not always show the best way to do something in the first place.
Typically, the simplest remedy is used.
Basic Meaning
The message "Object not set to an instance of Object" means you are trying to use an object which has not been initialized. This boils down to one of these:
Your code declared an object variable, but it did not initialize it (create an instance or 'instantiate' it)
Something which your code assumed would initialize an object, did not
Possibly, other code prematurely invalidated an object still in use
Finding The Cause
Since the problem is an object reference which is Nothing, the answer is to examine them to find out which one. Then determine why it is not initialized. Hold the mouse over the various variables and Visual Studio (VS) will show their values - the culprit will be Nothing.
You should also remove any Try/Catch blocks from the relevant code, especially ones where there is nothing in the Catch block. This will cause your code to crash when it tries to use an object which is Nothing. This is what you want because it will identify the exact location of the problem, and allow you to identify the object causing it.
A MsgBox in the Catch which displays Error while... will be of little help. This method also leads to very bad Stack  Overflow questions, because you can't describe the actual exception, the object involved or even the line of code where it happens.
You can also use the Locals Window (Debug -> Windows -> Locals) to examine your objects.
Once you know what and where the problem is, it is usually fairly easy to fix and faster than posting a new question.
See also:
Breakpoints
MSDN: How to: Use the Try/Catch Block to Catch Exceptions
MSDN: Best Practices for Exceptions
Examples and Remedies
Class Objects / Creating an Instance
Dim reg As CashRegister
...
TextBox1.Text = reg.Amount ' NRE
The problem is that Dim does not create a CashRegister object; it only declares a variable named reg of that Type. Declaring an object variable and creating an instance are two different things.
Remedy
The New operator can often be used to create the instance when you declare it:
Dim reg As New CashRegister ' [New] creates instance, invokes the constructor
' Longer, more explicit form:
Dim reg As CashRegister = New CashRegister
When it is only appropriate to create the instance later:
Private reg As CashRegister ' Declare
...
reg = New CashRegister() ' Create instance
Note: Do not use Dim again in a procedure, including the constructor (Sub New):
Private reg As CashRegister
'...
Public Sub New()
'...
Dim reg As New CashRegister
End Sub
This will create a local variable, reg, which exists only in that context (sub). The reg variable with module level Scope which you will use everywhere else remains Nothing.
Missing the New operator is the #1 cause of NullReference Exceptions seen in the Stack  Overflow questions reviewed.
Visual Basic tries to make the process clear repeatedly using New: Using the New Operator creates a new object and calls Sub New -- the constructor -- where your object can perform any other initialization.
To be clear, Dim (or Private) only declares a variable and its Type. The Scope of the variable - whether it exists for the entire module/class or is local to a procedure - is determined by where it is declared. Private | Friend | Public defines the access level, not Scope.
For more information, see:
New Operator
Scope in Visual Basic
Access Levels in Visual Basic
Value Types and Reference Types
Arrays
Arrays must also be instantiated:
Private arr as String()
This array has only been declared, not created. There are several ways to initialize an array:
Private arr as String() = New String(10){}
' or
Private arr() As String = New String(10){}
' For a local array (in a procedure) and using 'Option Infer':
Dim arr = New String(10) {}
Note: Beginning with VS 2010, when initializing a local array using a literal and Option Infer, the As <Type> and New elements are optional:
Dim myDbl As Double() = {1.5, 2, 9.9, 18, 3.14}
Dim myDbl = New Double() {1.5, 2, 9.9, 18, 3.14}
Dim myDbl() = {1.5, 2, 9.9, 18, 3.14}
The data Type and array size are inferred from the data being assigned. Class/Module level declarations still require As <Type> with Option Strict:
Private myDoubles As Double() = {1.5, 2, 9.9, 18, 3.14}
Example: Array of class objects
Dim arrFoo(5) As Foo
For i As Integer = 0 To arrFoo.Count - 1
arrFoo(i).Bar = i * 10 ' Exception
Next
The array has been created, but the Foo objects in it have not.
Remedy
For i As Integer = 0 To arrFoo.Count - 1
arrFoo(i) = New Foo() ' Create Foo instance
arrFoo(i).Bar = i * 10
Next
Using a List(Of T) will make it quite difficult to have an element without a valid object:
Dim FooList As New List(Of Foo) ' List created, but it is empty
Dim f As Foo ' Temporary variable for the loop
For i As Integer = 0 To 5
f = New Foo() ' Foo instance created
f.Bar = i * 10
FooList.Add(f) ' Foo object added to list
Next
For more information, see:
Option Infer Statement
Scope in Visual Basic
Arrays in Visual Basic
Lists and Collections
.NET collections (of which there are many varieties - Lists, Dictionary, etc.) must also be instantiated or created.
Private myList As List(Of String)
..
myList.Add("ziggy") ' NullReference
You get the same exception for the same reason - myList was only declared, but no instance created. The remedy is the same:
myList = New List(Of String)
' Or create an instance when declared:
Private myList As New List(Of String)
A common oversight is a class which uses a collection Type:
Public Class Foo
Private barList As List(Of Bar)
Friend Function BarCount As Integer
Return barList.Count
End Function
Friend Sub AddItem(newBar As Bar)
If barList.Contains(newBar) = False Then
barList.Add(newBar)
End If
End Function
Either procedure will result in an NRE, because barList is only declared, not instantiated. Creating an instance of Foo will not also create an instance of the internal barList. It may have been the intent to do this in the constructor:
Public Sub New ' Constructor
' Stuff to do when a new Foo is created...
barList = New List(Of Bar)
End Sub
As before, this is incorrect:
Public Sub New()
' Creates another barList local to this procedure
Dim barList As New List(Of Bar)
End Sub
For more information, see List(Of T) Class.
Data Provider Objects
Working with databases presents many opportunities for a NullReference because there can be many objects (Command, Connection, Transaction, Dataset, DataTable, DataRows....) in use at once. Note: It does not matter which data provider you are using -- MySQL, SQL Server, OleDB, etc. -- the concepts are the same.
Example 1
Dim da As OleDbDataAdapter
Dim ds As DataSet
Dim MaxRows As Integer
con.Open()
Dim sql = "SELECT * FROM tblfoobar_List"
da = New OleDbDataAdapter(sql, con)
da.Fill(ds, "foobar")
con.Close()
MaxRows = ds.Tables("foobar").Rows.Count ' Error
As before, the ds Dataset object was declared, but an instance was never created. The DataAdapter will fill an existing DataSet, not create one. In this case, since ds is a local variable, the IDE warns you that this might happen:
When declared as a module/class level variable, as appears to be the case with con, the compiler can't know if the object was created by an upstream procedure. Do not ignore warnings.
Remedy
Dim ds As New DataSet
Example 2
ds = New DataSet
da = New OleDBDataAdapter(sql, con)
da.Fill(ds, "Employees")
txtID.Text = ds.Tables("Employee").Rows(0).Item(1)
txtID.Name = ds.Tables("Employee").Rows(0).Item(2)
A typo is a problem here: Employees vs Employee. There was no DataTable named "Employee" created, so a NullReferenceException results trying to access it. Another potential problem is assuming there will be Items which may not be so when the SQL includes a WHERE clause.
Remedy
Since this uses one table, using Tables(0) will avoid spelling errors. Examining Rows.Count can also help:
If ds.Tables(0).Rows.Count > 0 Then
txtID.Text = ds.Tables(0).Rows(0).Item(1)
txtID.Name = ds.Tables(0).Rows(0).Item(2)
End If
Fill is a function returning the number of Rows affected which can also be tested:
If da.Fill(ds, "Employees") > 0 Then...
Example 3
Dim da As New OleDb.OleDbDataAdapter("SELECT TICKET.TICKET_NO,
TICKET.CUSTOMER_ID, ... FROM TICKET_RESERVATION AS TICKET INNER JOIN
FLIGHT_DETAILS AS FLIGHT ... WHERE [TICKET.TICKET_NO]= ...", con)
Dim ds As New DataSet
da.Fill(ds)
If ds.Tables("TICKET_RESERVATION").Rows.Count > 0 Then
The DataAdapter will provide TableNames as shown in the previous example, but it does not parse names from the SQL or database table. As a result, ds.Tables("TICKET_RESERVATION") references a non-existent table.
The Remedy is the same, reference the table by index:
If ds.Tables(0).Rows.Count > 0 Then
See also DataTable Class.
Object Paths / Nested
If myFoo.Bar.Items IsNot Nothing Then
...
The code is only testing Items while both myFoo and Bar may also be Nothing. The remedy is to test the entire chain or path of objects one at a time:
If (myFoo IsNot Nothing) AndAlso
(myFoo.Bar IsNot Nothing) AndAlso
(myFoo.Bar.Items IsNot Nothing) Then
....
AndAlso is important. Subsequent tests will not be performed once the first False condition is encountered. This allows the code to safely 'drill' into the object(s) one 'level' at a time, evaluating myFoo.Bar only after (and if) myFoo is determined to be valid. Object chains or paths can get quite long when coding complex objects:
myBase.myNodes(3).Layer.SubLayer.Foo.Files.Add("somefilename")
It is not possible to reference anything 'downstream' of a null object. This also applies to controls:
myWebBrowser.Document.GetElementById("formfld1").InnerText = "some value"
Here, myWebBrowser or Document could be Nothing or the formfld1 element may not exist.
UI Controls
Dim cmd5 As New SqlCommand("select Cartons, Pieces, Foobar " _
& "FROM Invoice where invoice_no = '" & _
Me.ComboBox5.SelectedItem.ToString.Trim & "' And category = '" & _
Me.ListBox1.SelectedItem.ToString.Trim & "' And item_name = '" & _
Me.ComboBox2.SelectedValue.ToString.Trim & "' And expiry_date = '" & _
Me.expiry.Text & "'", con)
Among other things, this code does not anticipate that the user may not have selected something in one or more UI controls. ListBox1.SelectedItem may well be Nothing, so ListBox1.SelectedItem.ToString will result in an NRE.
Remedy
Validate data before using it (also use Option Strict and SQL parameters):
Dim expiry As DateTime ' for text date validation
If (ComboBox5.SelectedItems.Count > 0) AndAlso
(ListBox1.SelectedItems.Count > 0) AndAlso
(ComboBox2.SelectedItems.Count > 0) AndAlso
(DateTime.TryParse(expiry.Text, expiry) Then
'... do stuff
Else
MessageBox.Show(...error message...)
End If
Alternatively, you can use (ComboBox5.SelectedItem IsNot Nothing) AndAlso...
Visual Basic Forms
Public Class Form1
Private NameBoxes = New TextBox(5) {Controls("TextBox1"), _
Controls("TextBox2"), Controls("TextBox3"), _
Controls("TextBox4"), Controls("TextBox5"), _
Controls("TextBox6")}
' same thing in a different format:
Private boxList As New List(Of TextBox) From {TextBox1, TextBox2, TextBox3 ...}
' Immediate NRE:
Private somevar As String = Me.Controls("TextBox1").Text
This is a fairly common way to get an NRE. In C#, depending on how it is coded, the IDE will report that Controls does not exist in the current context, or "cannot reference non-static member". So, to some extent, this is a VB-only situation. It is also complex because it can result in a failure cascade.
The arrays and collections cannot be initialized this way. This initialization code will run before the constructor creates the Form or the Controls. As a result:
Lists and Collection will simply be empty
The Array will contain five elements of Nothing
The somevar assignment will result in an immediate NRE because Nothing doesn't have a .Text property
Referencing array elements later will result in an NRE. If you do this in Form_Load, due to an odd bug, the IDE may not report the exception when it happens. The exception will pop up later when your code tries to use the array. This "silent exception" is detailed in this post. For our purposes, the key is that when something catastrophic happens while creating a form (Sub New or Form Load event), exceptions may go unreported, the code exits the procedure and just displays the form.
Since no other code in your Sub New or Form Load event will run after the NRE, a great many other things can be left uninitialized.
Sub Form_Load(..._
'...
Dim name As String = NameBoxes(2).Text ' NRE
' ...
' More code (which will likely not be executed)
' ...
End Sub
Note this applies to any and all control and component references making these illegal where they are:
Public Class Form1
Private myFiles() As String = Me.OpenFileDialog1.FileName & ...
Private dbcon As String = OpenFileDialog1.FileName & ";Jet Oledb..."
Private studentName As String = TextBox13.Text
Partial Remedy
It is curious that VB does not provide a warning, but the remedy is to declare the containers at the form level, but initialize them in form load event handler when the controls do exist. This can be done in Sub New as long as your code is after the InitializeComponent call:
' Module level declaration
Private NameBoxes as TextBox()
Private studentName As String
' Form Load, Form Shown or Sub New:
'
' Using the OP's approach (illegal using OPTION STRICT)
NameBoxes = New TextBox() {Me.Controls("TextBox1"), Me.Controls("TestBox2"), ...)
studentName = TextBox32.Text ' For simple control references
The array code may not be out of the woods yet. Any controls which are in a container control (like a GroupBox or Panel) will not be found in Me.Controls; they will be in the Controls collection of that Panel or GroupBox. Nor will a control be returned when the control name is misspelled ("TeStBox2"). In such cases, Nothing will again be stored in those array elements and an NRE will result when you attempt to reference it.
These should be easy to find now that you know what you are looking for:
"Button2" resides on a Panel
Remedy
Rather than indirect references by name using the form's Controls collection, use the control reference:
' Declaration
Private NameBoxes As TextBox()
' Initialization - simple and easy to read, hard to botch:
NameBoxes = New TextBox() {TextBox1, TextBox2, ...)
' Initialize a List
NamesList = New List(Of TextBox)({TextBox1, TextBox2, TextBox3...})
' or
NamesList = New List(Of TextBox)
NamesList.AddRange({TextBox1, TextBox2, TextBox3...})
Function Returning Nothing
Private bars As New List(Of Bars) ' Declared and created
Public Function BarList() As List(Of Bars)
bars.Clear
If someCondition Then
For n As Integer = 0 to someValue
bars.Add(GetBar(n))
Next n
Else
Exit Function
End If
Return bars
End Function
This is a case where the IDE will warn you that 'not all paths return a value and a NullReferenceException may result'. You can suppress the warning, by replacing Exit Function with Return Nothing, but that does not solve the problem. Anything which tries to use the return when someCondition = False will result in an NRE:
bList = myFoo.BarList()
For Each b As Bar in bList ' EXCEPTION
...
Remedy
Replace Exit Function in the function with Return bList. Returning an empty List is not the same as returning Nothing. If there is a chance that a returned object can be Nothing, test before using it:
bList = myFoo.BarList()
If bList IsNot Nothing Then...
Poorly Implemented Try/Catch
A badly implemented Try/Catch can hide where the problem is and result in new ones:
Dim dr As SqlDataReader
Try
Dim lnk As LinkButton = TryCast(sender, LinkButton)
Dim gr As GridViewRow = DirectCast(lnk.NamingContainer, GridViewRow)
Dim eid As String = GridView1.DataKeys(gr.RowIndex).Value.ToString()
ViewState("username") = eid
sqlQry = "select FirstName, Surname, DepartmentName, ExtensionName, jobTitle,
Pager, mailaddress, from employees1 where username='" & eid & "'"
If connection.State <> ConnectionState.Open Then
connection.Open()
End If
command = New SqlCommand(sqlQry, connection)
'More code fooing and barring
dr = command.ExecuteReader()
If dr.Read() Then
lblFirstName.Text = Convert.ToString(dr("FirstName"))
...
End If
mpe.Show()
Catch
Finally
command.Dispose()
dr.Close() ' <-- NRE
connection.Close()
End Try
This is a case of an object not being created as expected, but also demonstrates the counter usefulness of an empty Catch.
There is an extra comma in the SQL (after 'mailaddress') which results in an exception at .ExecuteReader. After the Catch does nothing, Finally tries to perform clean up, but since you cannot Close a null DataReader object, a brand new NullReferenceException results.
An empty Catch block is the devil's playground. This OP was baffled why he was getting an NRE in the Finally block. In other situations, an empty Catch may result in something else much further downstream going haywire and cause you to spend time looking at the wrong things in the wrong place for the problem. (The "silent exception" described above provides the same entertainment value.)
Remedy
Don't use empty Try/Catch blocks - let the code crash so you can a) identify the cause b) identify the location and c) apply a proper remedy. Try/Catch blocks are not intended to hide exceptions from the person uniquely qualified to fix them - the developer.
DBNull is not the same as Nothing
For Each row As DataGridViewRow In dgvPlanning.Rows
If Not IsDBNull(row.Cells(0).Value) Then
...
The IsDBNull function is used to test if a value equals System.DBNull: From MSDN:
The System.DBNull value indicates that the Object represents missing or non-existent data. DBNull is not the same as Nothing, which indicates that a variable has not yet been initialized.
Remedy
If row.Cells(0) IsNot Nothing Then ...
As before, you can test for Nothing, then for a specific value:
If (row.Cells(0) IsNot Nothing) AndAlso (IsDBNull(row.Cells(0).Value) = False) Then
Example 2
Dim getFoo = (From f In dbContext.FooBars
Where f.something = something
Select f).FirstOrDefault
If Not IsDBNull(getFoo) Then
If IsDBNull(getFoo.user_id) Then
txtFirst.Text = getFoo.first_name
Else
...
FirstOrDefault returns the first item or the default value, which is Nothing for reference types and never DBNull:
If getFoo IsNot Nothing Then...
Controls
Dim chk As CheckBox
chk = CType(Me.Controls(chkName), CheckBox)
If chk.Checked Then
Return chk
End If
If a CheckBox with chkName can't be found (or exists in a GroupBox), then chk will be Nothing and be attempting to reference any property will result in an exception.
Remedy
If (chk IsNot Nothing) AndAlso (chk.Checked) Then ...
The DataGridView
The DGV has a few quirks seen periodically:
dgvBooks.DataSource = loan.Books
dgvBooks.Columns("ISBN").Visible = True ' NullReferenceException
dgvBooks.Columns("Title").DefaultCellStyle.Format = "C"
dgvBooks.Columns("Author").DefaultCellStyle.Format = "C"
dgvBooks.Columns("Price").DefaultCellStyle.Format = "C"
If dgvBooks has AutoGenerateColumns = True, it will create the columns, but it does not name them, so the above code fails when it references them by name.
Remedy
Name the columns manually, or reference by index:
dgvBooks.Columns(0).Visible = True
Example 2 — Beware of the NewRow
xlWorkSheet = xlWorkBook.Sheets("sheet1")
For i = 0 To myDGV.RowCount - 1
For j = 0 To myDGV.ColumnCount - 1
For k As Integer = 1 To myDGV.Columns.Count
xlWorkSheet.Cells(1, k) = myDGV.Columns(k - 1).HeaderText
xlWorkSheet.Cells(i + 2, j + 1) = myDGV(j, i).Value.ToString()
Next
Next
Next
When your DataGridView has AllowUserToAddRows as True (the default), the Cells in the blank/new row at the bottom will all contain Nothing. Most attempts to use the contents (for example, ToString) will result in an NRE.
Remedy
Use a For/Each loop and test the IsNewRow property to determine if it is that last row. This works whether AllowUserToAddRows is true or not:
For Each r As DataGridViewRow in myDGV.Rows
If r.IsNewRow = False Then
' ok to use this row
If you do use a For n loop, modify the row count or use Exit For when IsNewRow is true.
My.Settings (StringCollection)
Under certain circumstances, trying to use an item from My.Settings which is a StringCollection can result in a NullReference the first time you use it. The solution is the same, but not as obvious. Consider:
My.Settings.FooBars.Add("ziggy") ' foobars is a string collection
Since VB is managing Settings for you, it is reasonable to expect it to initialize the collection. It will, but only if you have previously added an initial entry to the collection (in the Settings editor). Since the collection is (apparently) initialized when an item is added, it remains Nothing when there are no items in the Settings editor to add.
Remedy
Initialize the settings collection in the form's Load event handler, if/when needed:
If My.Settings.FooBars Is Nothing Then
My.Settings.FooBars = New System.Collections.Specialized.StringCollection
End If
Typically, the Settings collection will only need to be initialized the first time the application runs. An alternate remedy is to add an initial value to your collection in Project -> Settings | FooBars, save the project, then remove the fake value.
Key Points
You probably forgot the New operator.
or
Something you assumed would perform flawlessly to return an initialized object to your code, did not.
Don't ignore compiler warnings (ever) and use Option Strict On (always).
MSDN NullReference Exception
Another scenario is when you cast a null object into a value type. For example, the code below:
object o = null;
DateTime d = (DateTime)o;
It will throw a NullReferenceException on the cast. It seems quite obvious in the above sample, but this can happen in more "late-binding" intricate scenarios where the null object has been returned from some code you don't own, and the cast is for example generated by some automatic system.
One example of this is this simple ASP.NET binding fragment with the Calendar control:
<asp:Calendar runat="server" SelectedDate="<%#Bind("Something")%>" />
Here, SelectedDate is in fact a property - of DateTime type - of the Calendar Web Control type, and the binding could perfectly return something null. The implicit ASP.NET Generator will create a piece of code that will be equivalent to the cast code above. And this will raise a NullReferenceException that is quite difficult to spot, because it lies in ASP.NET generated code which compiles fine...
It means your code used an object reference variable that was set to null (i.e. it did not reference an actual object instance).
To prevent the error, objects that could be null should be tested for null before being used.
if (myvar != null)
{
// Go ahead and use myvar
myvar.property = ...
}
else
{
// Whoops! myvar is null and cannot be used without first
// assigning it to an instance reference
// Attempting to use myvar here will result in NullReferenceException
}
It means that the variable in question is pointed at nothing. I could generate this like so:
SqlConnection connection = null;
connection.Open();
That will throw the error because while I've declared the variable "connection", it's not pointed to anything. When I try to call the member "Open", there's no reference for it to resolve, and it will throw the error.
To avoid this error:
Always initialize your objects before you try to do anything with them.
If you're not sure whether the object is null, check it with object == null.
JetBrains' ReSharper tool will identify every place in your code that has the possibility of a null reference error, allowing you to put in a null check. This error is the number one source of bugs, IMHO.
Be aware that regardless of the scenario, the cause is always the same in .NET:
You are trying to use a reference variable whose value is Nothing/null. When the value is Nothing/null for the reference variable, that means it is not actually holding a reference to an instance of any object that exists on the heap.
You either never assigned something to the variable, never created an instance of the value assigned to the variable, or you set the variable equal to Nothing/null manually, or you called a function that set the variable to Nothing/null for you.
An example of this exception being thrown is: When you are trying to check something, that is null.
For example:
string testString = null; //Because it doesn't have a value (i.e. it's null; "Length" cannot do what it needs to do)
if (testString.Length == 0) // Throws a nullreferenceexception
{
//Do something
}
The .NET runtime will throw a NullReferenceException when you attempt to perform an action on something which hasn't been instantiated i.e. the code above.
In comparison to an ArgumentNullException which is typically thrown as a defensive measure if a method expects that what is being passed to it is not null.
More information is in C# NullReferenceException and Null Parameter.
Update C#8.0, 2019: Nullable reference types
C#8.0 introduces nullable reference types and non-nullable reference types. So only nullable reference types must be checked to avoid a NullReferenceException.
If you have not initialized a reference type, and you want to set or read one of its properties, it will throw a NullReferenceException.
Example:
Person p = null;
p.Name = "Harry"; // NullReferenceException occurs here.
You can simply avoid this by checking if the variable is not null:
Person p = null;
if (p!=null)
{
p.Name = "Harry"; // Not going to run to this point
}
To fully understand why a NullReferenceException is thrown, it is important to know the difference between value types and [reference types][3].
So, if you're dealing with value types, NullReferenceExceptions can not occur. Though you need to keep alert when dealing with reference types!
Only reference types, as the name is suggesting, can hold references or point literally to nothing (or 'null'). Whereas value types always contain a value.
Reference types (these ones must be checked):
dynamic
object
string
Value types (you can simply ignore these ones):
Numeric types
Integral types
Floating-point types
decimal
bool
User defined structs
Another case where NullReferenceExceptions can happen is the (incorrect) use of the as operator:
class Book {
public string Name { get; set; }
}
class Car { }
Car mycar = new Car();
Book mybook = mycar as Book; // Incompatible conversion --> mybook = null
Console.WriteLine(mybook.Name); // NullReferenceException
Here, Book and Car are incompatible types; a Car cannot be converted/cast to a Book. When this cast fails, as returns null. Using mybook after this causes a NullReferenceException.
In general, you should use a cast or as, as follows:
If you are expecting the type conversion to always succeed (ie. you know what the object should be ahead of time), then you should use a cast:
ComicBook cb = (ComicBook)specificBook;
If you are unsure of the type, but you want to try to use it as a specific type, then use as:
ComicBook cb = specificBook as ComicBook;
if (cb != null) {
// ...
}
You are using the object that contains the null value reference. So it's giving a null exception. In the example the string value is null and when checking its length, the exception occurred.
Example:
string value = null;
if (value.Length == 0) // <-- Causes exception
{
Console.WriteLine(value); // <-- Never reached
}
The exception error is:
Unhandled Exception:
System.NullReferenceException: Object reference not set to an instance
of an object. at Program.Main()
While what causes a NullReferenceExceptions and approaches to avoid/fix such an exception have been addressed in other answers, what many programmers haven't learned yet is how to independently debug such exceptions during development.
In Visual Studio this is usually easy thanks to the Visual Studio Debugger.
First, make sure that the correct error is going to be caught - see
How do I allow breaking on 'System.NullReferenceException' in VS2010? Note1
Then either Start with Debugging (F5) or Attach [the VS Debugger] to Running Process. On occasion it may be useful to use Debugger.Break, which will prompt to launch the debugger.
Now, when the NullReferenceException is thrown (or unhandled) the debugger will stop (remember the rule set above?) on the line on which the exception occurred. Sometimes the error will be easy to spot.
For instance,
in the following line the only code that can cause the exception is if myString evaluates to null. This can be verified by looking at the Watch Window or running expressions in the Immediate Window.
var x = myString.Trim();
In more advanced cases, such as the following, you'll need to use one of the techniques above (Watch or Immediate Windows) to inspect the expressions to determine if str1 was null or if str2 was null.
var x = str1.Trim() + str2.Trim();
Once where the exception is throw has been located, it's usually trivial to reason backwards to find out where the null value was [incorrectly] introduced --
Take the time required to understand the cause of the exception. Inspect for null expressions. Inspect the previous expressions which could have resulted in such null expressions. Add breakpoints and step through the program as appropriate. Use the debugger.
1 If Break on Throws is too aggressive and the debugger stops on an NPE in the .NET or 3rd-party library, Break on User-Unhandled can be used to limit the exceptions caught. Additionally, VS2012 introduces Just My Code which I recommend enabling as well.
If you are debugging with Just My Code enabled, the behavior is slightly different. With Just My Code enabled, the debugger ignores first-chance common language runtime (CLR) exceptions that are thrown outside of My Code and do not pass through My Code
Simon Mourier gave this example:
object o = null;
DateTime d = (DateTime)o; // NullReferenceException
where an unboxing conversion (cast) from object (or from one of the classes System.ValueType or System.Enum, or from an interface type) to a value type (other than Nullable<>) in itself gives the NullReferenceException.
In the other direction, a boxing conversion from a Nullable<> which has HasValue equal to false to a reference type, can give a null reference which can then later lead to a NullReferenceException. The classic example is:
DateTime? d = null;
var s = d.ToString(); // OK, no exception (no boxing), returns ""
var t = d.GetType(); // Bang! d is boxed, NullReferenceException
Sometimes the boxing happens in another way. For example with this non-generic extension method:
public static void MyExtension(this object x)
{
x.ToString();
}
the following code will be problematic:
DateTime? d = null;
d.MyExtension(); // Leads to boxing, NullReferenceException occurs inside the body of the called method, not here.
These cases arise because of the special rules the runtime uses when boxing Nullable<> instances.
Adding a case when the class name for entity used in entity framework is same as class name for a web form code-behind file.
Suppose you have a web form Contact.aspx whose codebehind class is Contact and you have an entity name Contact.
Then following code will throw a NullReferenceException when you call context.SaveChanges()
Contact contact = new Contact { Name = "Abhinav"};
var context = new DataContext();
context.Contacts.Add(contact);
context.SaveChanges(); // NullReferenceException at this line
For the sake of completeness DataContext class
public class DataContext : DbContext
{
public DbSet<Contact> Contacts {get; set;}
}
and Contact entity class. Sometimes entity classes are partial classes so that you can extend them in other files too.
public partial class Contact
{
public string Name {get; set;}
}
The error occurs when both the entity and codebehind class are in same namespace.
To fix this, rename the entity class or the codebehind class for Contact.aspx.
Reason
I am still not sure about the reason. But whenever any of the entity class will extend System.Web.UI.Page this error occurs.
For discussion have a look at NullReferenceException in DbContext.saveChanges()
Another general case where one might receive this exception involves mocking classes during unit testing. Regardless of the mocking framework being used, you must ensure that all appropriate levels of the class hierarchy are properly mocked. In particular, all properties of HttpContext which are referenced by the code under test must be mocked.
See "NullReferenceException thrown when testing custom AuthorizationAttribute" for a somewhat verbose example.
I have a different perspective to answering this. This sort of answers "what else can I do to avoid it?"
When working across different layers, for example in an MVC application, a controller needs services to call business operations. In such scenarios Dependency Injection Container can be used to initialize the services to avoid the NullReferenceException. So that means you don't need to worry about checking for null and just call the services from the controller as though they will always to available (and initialized) as either a singleton or a prototype.
public class MyController
{
private ServiceA serviceA;
private ServiceB serviceB;
public MyController(ServiceA serviceA, ServiceB serviceB)
{
this.serviceA = serviceA;
this.serviceB = serviceB;
}
public void MyMethod()
{
// We don't need to check null because the dependency injection container
// injects it, provided you took care of bootstrapping it.
var someObject = serviceA.DoThis();
}
}
On the matter of "what should I do about it", there can be many answers.
A more "formal" way of preventing such error conditions while developing is applying design by contract in your code. This means you need to set class invariants, and/or even function/method preconditions and postconditions on your system, while developing.
In short, class invariants ensure that there will be some constraints in your class that will not get violated in normal use (and therefore, the class will not get in an inconsistent state). Preconditions mean that data given as input to a function/method must follow some constraints set and never violate them, and postconditions mean that a function/method output must follow the set constraints again without ever violating them.
Contract conditions should never be violated during execution of a bug-free program, therefore design by contract is checked in practice in debug mode, while being disabled in releases, to maximize the developed system performance.
This way, you can avoid NullReferenceException cases that are results of violation of the constraints set. For example, if you use an object property X in a class and later try to invoke one of its methods and X has a null value, then this will lead to NullReferenceException:
public X { get; set; }
public void InvokeX()
{
X.DoSomething(); // if X value is null, you will get a NullReferenceException
}
But if you set "property X must never have a null value" as method precondition, then you can prevent the scenario described before:
//Using code contracts:
[ContractInvariantMethod]
protected void ObjectInvariant()
{
Contract.Invariant(X != null);
//...
}
For this cause, Code Contracts project exists for .NET applications.
Alternatively, design by contract can be applied using assertions.
UPDATE: It is worth mentioning that the term was coined by Bertrand Meyer in connection with his design of the Eiffel programming language.
A NullReferenceException is thrown when we are trying to access Properties of a null object or when a string value becomes empty and we are trying to access string methods.
For example:
When a string method of an empty string accessed:
string str = string.Empty;
str.ToLower(); // throw null reference exception
When a property of a null object accessed:
Public Class Person {
public string Name { get; set; }
}
Person objPerson;
objPerson.Name /// throw Null refernce Exception
TL;DR: Try using Html.Partial instead of Renderpage
I was getting Object reference not set to an instance of an object when I tried to render a View within a View by sending it a Model, like this:
#{
MyEntity M = new MyEntity();
}
#RenderPage("_MyOtherView.cshtml", M); // error in _MyOtherView, the Model was Null
Debugging showed the model was Null inside MyOtherView. Until I changed it to:
#{
MyEntity M = new MyEntity();
}
#Html.Partial("_MyOtherView.cshtml", M);
And it worked.
Furthermore, the reason I didn't have Html.Partial to begin with was because Visual Studio sometimes throws error-looking squiggly lines under Html.Partial if it's inside a differently constructed foreach loop, even though it's not really an error:
#inherits System.Web.Mvc.WebViewPage
#{
ViewBag.Title = "Entity Index";
List<MyEntity> MyEntities = new List<MyEntity>();
MyEntities.Add(new MyEntity());
MyEntities.Add(new MyEntity());
MyEntities.Add(new MyEntity());
}
<div>
#{
foreach(var M in MyEntities)
{
// Squiggly lines below. Hovering says: cannot convert method group 'partial' to non-delegate type Object, did you intend to envoke the Method?
#Html.Partial("MyOtherView.cshtml");
}
}
</div>
But I was able to run the application with no problems with this "error". I was able to get rid of the error by changing the structure of the foreach loop to look like this:
#foreach(var M in MyEntities){
...
}
Although I have a feeling it was because Visual Studio was misreading the ampersands and brackets.
What can you do about it?
There is a lot of good answers here explaining what a null reference is and how to debug it. But there is very little on how to prevent the issue or at least make it easier to catch.
Check arguments
For example, methods can check the different arguments to see if they are null and throw an ArgumentNullException, an exception obviously created for this exact purpose.
The constructor for the ArgumentNullException even takes the name of the parameter and a message as arguments so you can tell the developer exactly what the problem is.
public void DoSomething(MyObject obj) {
if(obj == null)
{
throw new ArgumentNullException("obj", "Need a reference to obj.");
}
}
Use Tools
There are also several libraries that can help. "Resharper" for example can provide you with warnings while you are writing code, especially if you use their attribute: NotNullAttribute
There's "Microsoft Code Contracts" where you use syntax like Contract.Requires(obj != null) which gives you runtime and compile checking: Introducing Code Contracts.
There's also "PostSharp" which will allow you to just use attributes like this:
public void DoSometing([NotNull] obj)
By doing that and making PostSharp part of your build process obj will be checked for null at runtime. See: PostSharp null check
Plain Code Solution
Or you can always code your own approach using plain old code. For example here is a struct that you can use to catch null references. It's modeled after the same concept as Nullable<T>:
[System.Diagnostics.DebuggerNonUserCode]
public struct NotNull<T> where T: class
{
private T _value;
public T Value
{
get
{
if (_value == null)
{
throw new Exception("null value not allowed");
}
return _value;
}
set
{
if (value == null)
{
throw new Exception("null value not allowed.");
}
_value = value;
}
}
public static implicit operator T(NotNull<T> notNullValue)
{
return notNullValue.Value;
}
public static implicit operator NotNull<T>(T value)
{
return new NotNull<T> { Value = value };
}
}
You would use very similar to the same way you would use Nullable<T>, except with the goal of accomplishing exactly the opposite - to not allow null. Here are some examples:
NotNull<Person> person = null; // throws exception
NotNull<Person> person = new Person(); // OK
NotNull<Person> person = GetPerson(); // throws exception if GetPerson() returns null
NotNull<T> is implicitly cast to and from T so you can use it just about anywhere you need it. For example, you can pass a Person object to a method that takes a NotNull<Person>:
Person person = new Person { Name = "John" };
WriteName(person);
public static void WriteName(NotNull<Person> person)
{
Console.WriteLine(person.Value.Name);
}
As you can see above as with nullable you would access the underlying value through the Value property. Alternatively, you can use an explicit or implicit cast, you can see an example with the return value below:
Person person = GetPerson();
public static NotNull<Person> GetPerson()
{
return new Person { Name = "John" };
}
Or you can even use it when the method just returns T (in this case Person) by doing a cast. For example, the following code would just like the code above:
Person person = (NotNull<Person>)GetPerson();
public static Person GetPerson()
{
return new Person { Name = "John" };
}
Combine with Extension
Combine NotNull<T> with an extension method and you can cover even more situations. Here is an example of what the extension method can look like:
[System.Diagnostics.DebuggerNonUserCode]
public static class NotNullExtension
{
public static T NotNull<T>(this T #this) where T: class
{
if (#this == null)
{
throw new Exception("null value not allowed");
}
return #this;
}
}
And here is an example of how it could be used:
var person = GetPerson().NotNull();
GitHub
For your reference I made the code above available on GitHub, you can find it at:
https://github.com/luisperezphd/NotNull
Related Language Feature
C# 6.0 introduced the "null-conditional operator" that helps with this a little. With this feature, you can reference nested objects and if any one of them is null the whole expression returns null.
This reduces the number of null checks you have to do in some cases. The syntax is to put a question mark before each dot. Take the following code for example:
var address = country?.State?.County?.City;
Imagine that country is an object of type Country that has a property called State and so on. If country, State, County, or City is null then address will benull. Therefore you only have to check whetheraddressisnull`.
It's a great feature, but it gives you less information. It doesn't make it obvious which of the 4 is null.
Built-in like Nullable?
C# has a nice shorthand for Nullable<T>, you can make something nullable by putting a question mark after the type like so int?.
It would be nice if C# had something like the NotNull<T> struct above and had a similar shorthand, maybe the exclamation point (!) so that you could write something like: public void WriteName(Person! person).
You can fix NullReferenceException in a clean way using Null-conditional Operators in C# 6 and write less code to handle null checks.
It's used to test for null before performing a member access (?.) or index (?[) operation.
Example
var name = p?.Spouse?.FirstName;
It is equivalent to:
if (p != null)
{
if (p.Spouse != null)
{
name = p.Spouse.FirstName;
}
}
The result is that the name will be null when p is null or when p.Spouse is null.
Otherwise, the variable name will be assigned the value of the p.Spouse.FirstName.
For more details: Null-conditional Operators
Interestingly, none of the answers on this page mention the two edge cases:
Edge case #1: concurrent access to a Dictionary
Generic dictionaries in .NET are not thread-safe and they sometimes might throw a NullReference or even (more frequent) a KeyNotFoundException when you try to access a key from two concurrent threads. The exception is quite misleading in this case.
Edge case #2: unsafe code
If a NullReferenceException is thrown by unsafe code, you might look at your pointer variables, and check them for IntPtr.Zero or something. Which is the same thing ("null pointer exception"), but in unsafe code, variables are often cast to value-types/arrays, etc., and you bang your head against the wall, wondering how a value-type can throw this exception.
(Another reason for non-using unsafe code unless you need it, by the way.)
Edge case #3: Visual Studio multi monitor setup with secondary monitor(s) that has different DPI setting than the primary monitor
This edge case is software-specific and pertains to the Visual Studio 2019 IDE (and possibly earlier versions).
A method to reproduce the problem: drag any component from the Toolbox to a Windows form on a non-primary monitor with different DPI setting than the primary monitor, and you get a pop-up with “Object reference not set to an instance of an object.” According to this thread, this issue has been known for quite some time and at the time of writing it still hasn't been fixed.
The error line "Object reference not set to an instance of an object." states that you have not assigned an instance object to a object reference and still you are accessing properties/methods of that object.
For example: let's say you have a class called myClass and it contains one property, prop1.
public Class myClass
{
public int prop1 {get;set;}
}
Now you are accessing this prop1 in some other class just like below:
public class Demo
{
public void testMethod()
{
myClass ref = null;
ref.prop1 = 1; // This line throws an error
}
}
The above line throws an error because reference of class myClass is declared, but not instantiated or an instance of object is not assigned to a reference of that class.
To fix this you have to instantiate (assign an object to a reference of that class).
public class Demo
{
public void testMethod()
{
myClass ref = null;
ref = new myClass();
ref.prop1 = 1;
}
}
NullReferenceException or Object reference not set to an instance of an object occurs when an object of the class you are trying to use is not instantiated.
For example:
Assume that you have a class named Student.
public class Student
{
private string FirstName;
private string LastName;
public string GetFullName()
{
return FirstName + LastName;
}
}
Now, consider another class where you are trying to retrieve the student's full name.
public class StudentInfo
{
public string GetStudentName()
{
Student s;
string fullname = s.GetFullName();
return fullname;
}
}
As seen in the above code, the statement
Student s - only declares the variable of type Student, note that the Student class is not instantiated at this point.
Hence, when the statement s.GetFullName() gets executed, it will throw the NullReferenceException.
Well, in simple terms:
You are trying to access an object that isn't created or currently not in memory.
So how to tackle this:
Debug and let the debugger break... It will directly take you to the variable that is broken... Now your task is to simply fix this.. Using the new keyword in the appropriate place.
If it is caused on some database commands because the object isn't present then all you need to do is do a null check and handle it:
if (i == null) {
// Handle this
}
The hardest one .. if the GC collected the object already... This generally occurs if you are trying to find an object using strings... That is, finding it by name of the object then it may happen that the GC might already cleaned it up... This is hard to find and will become quite a problem... A better way to tackle this is do null checks wherever necessary during the development process. This will save you a lot of time.
By finding by name I mean some framework allow you to FIndObjects using strings and the code might look like this: FindObject("ObjectName");
Literally the easiest way to fix a NullReferenceExeption has two ways.
If you have a GameObject for example with a script attached and a variable named rb (rigidbody) this variable will start with null when you start your game.
This is why you get a NullReferenceExeption because the computer does not have data stored in that variable.
I'll be using a RigidBody variable as an example.
We can add data really easily actually in a few ways:
Add a RigidBody to your object with AddComponent > Physics > Rigidbody
Then go into your script and type rb = GetComponent<Rigidbody>();
This line of code works best under your Start() or Awake() functions.
You can add a component programmatically and assign the variable at the same time with one line of code: rb = AddComponent<RigidBody>();
Further Notes: If you want Unity to add a component to your object and you might have forgotten to add one, you can type [RequireComponent(typeof(RigidBody))] above your class declaration (the space below all of your usings).
Enjoy and have fun making games!
If one is getting this message during saving or compiling the build, just close all the files and then open any file to compile and save.
For me the reason was that I had rename the file and the old file was still open.
This is basically is a Null reference exception. As Microsoft states-
A NullReferenceException exception is thrown when you try to access a
member of a type whose value is null.
What does that mean?
That means if any member which doesn’t hold any value and we are making that member to perform certain task then the system will undoubtedly toss a message and say-
“Hey wait, that member has no values so it can’t perform the task which you are handing it over.”
The exception itself says that something is being referred but whose value is not being set. So this denotes that it only occurs while using reference types as Value types are non-nullable.
NullReferenceException won't occur if we are using Value type members.
class Program
{
static void Main(string[] args)
{
string str = null;
Console.WriteLine(str.Length);
Console.ReadLine();
}
}
The above code shows simple string which is assigned with a null value.
Now, when I try to print the length of the string str, I do get An unhandled exception of type ‘System.NullReferenceException’ occurred message because member str is pointing to null and there can’t be any length of null.
‘NullReferenceException’ also occurs when we forget to instantiate a reference type.
Suppose I have a class and member method in it. I have not instantiated my class but only named my class. Now if I try to use the method, the compiler will throw an error or issue a warning (depending on the compiler).
class Program
{
static void Main(string[] args)
{
MyClass1 obj;
obj.foo(); // Use of unassigned local variable 'obj'
}
}
public class MyClass1
{
internal void foo()
{
Console.WriteLine("Hello from foo");
}
}
Compiler for the above code raises an error that variable obj is unassigned which signifies that our variable has null values or nothing. Compiler for the above code raises an error that variable obj is unassigned which signifies that our variable has null values or nothing.
Why does it occur?
NullReferenceException arises due to our fault for not checking the object’s value. We often leave the object values unchecked in the code development.
It also arises when we forget to instantiate our objects. Using methods, properties, collections etc. which can return or set null values can also be the cause of this exception.
How can it be avoided?
There are various ways and methods to avoid this renowned exception:
Explicit Checking: We should adhere to the tradition of checking the objects, properties, methods, arrays, and collections whether they are null. This can be simply implemented using conditional statements like if-else if-else etc.
Exception handling: One of the important ways of managing this exception. Using simple try-catch-finally blocks we can control this exception and also maintain a log of it. This can be very useful when your application is on production stage.
Null operators: Null Coalescing operator and null conditional operators can also be used in handy while setting values to objects, variables, properties and fields.
Debugger: For developers, we have the big weapon of Debugging with us. If have we face NullReferenceException during the development face we can use the debugger to get to the source of the exception.
Built-in method: System methods such as GetValueOrDefault(), IsNullOrWhiteSpace(), and IsNullorEmpty() checks for nulls and assign the default value if there is a null value.
There are many good answers already here. You can also check more detailed description with examples on my blog.
Hope this helps too!

Creating one parameter for multiple DB columns in ActiveRecord Rails [duplicate]

Ruby setters—whether created by (c)attr_accessor or manually—seem to be the only methods that need self. qualification when accessed within the class itself. This seems to put Ruby alone the world of languages:
All methods need self/this (like Perl, and I think Javascript)
No methods require self/this is (C#, Java)
Only setters need self/this (Ruby?)
The best comparison is C# vs Ruby, because both languages support accessor methods which work syntactically just like class instance variables: foo.x = y, y = foo.x . C# calls them properties.
Here's a simple example; the same program in Ruby then C#:
class A
def qwerty; #q; end # manual getter
def qwerty=(value); #q = value; end # manual setter, but attr_accessor is same
def asdf; self.qwerty = 4; end # "self." is necessary in ruby?
def xxx; asdf; end # we can invoke nonsetters w/o "self."
def dump; puts "qwerty = #{qwerty}"; end
end
a = A.new
a.xxx
a.dump
take away the self.qwerty =() and it fails (Ruby 1.8.6 on Linux & OS X). Now C#:
using System;
public class A {
public A() {}
int q;
public int qwerty {
get { return q; }
set { q = value; }
}
public void asdf() { qwerty = 4; } // C# setters work w/o "this."
public void xxx() { asdf(); } // are just like other methods
public void dump() { Console.WriteLine("qwerty = {0}", qwerty); }
}
public class Test {
public static void Main() {
A a = new A();
a.xxx();
a.dump();
}
}
Question: Is this true? Are there other occasions besides setters where self is necessary? I.e., are there other occasions where a Ruby method cannot be invoked without self?
There are certainly lots of cases where self becomes necessary. This is not unique to Ruby, just to be clear:
using System;
public class A {
public A() {}
public int test { get { return 4; }}
public int useVariable() {
int test = 5;
return test;
}
public int useMethod() {
int test = 5;
return this.test;
}
}
public class Test {
public static void Main() {
A a = new A();
Console.WriteLine("{0}", a.useVariable()); // prints 5
Console.WriteLine("{0}", a.useMethod()); // prints 4
}
}
Same ambiguity is resolved in same way. But while subtle I'm asking about the case where
A method has been defined, and
No local variable has been defined, and
we encounter
qwerty = 4
which is ambiguous—is this a method invocation or an new local variable assignment?
#Mike Stone
Hi! I understand and appreciate the points you've made and your
example was great. Believe me when I say, if I had enough reputation,
I'd vote up your response. Yet we still disagree:
on a matter of semantics, and
on a central point of fact
First I claim, not without irony, we're having a semantic debate about the
meaning of 'ambiguity'.
When it comes to parsing and programming language semantics (the subject
of this question), surely you would admit a broad spectrum of the notion
'ambiguity'. Let's just adopt some random notation:
ambiguous: lexical ambiguity (lex must 'look ahead')
Ambiguous: grammatical ambiguity (yacc must defer to parse-tree analysis)
AMBIGUOUS: ambiguity knowing everything at the moment of execution
(and there's junk between 2-3 too). All these categories are resolved by
gathering more contextual info, looking more and more globally. So when you
say,
"qwerty = 4" is UNAMBIGUOUS in C#
when there is no variable defined...
I couldn't agree more. But by the same token, I'm saying
"qwerty = 4" is un-Ambiguous in ruby
(as it now exists)
"qwerty = 4" is Ambiguous in C#
And we're not yet contradicting each other. Finally, here's where we really
disagree: Either ruby could or could not be implemented without any further
language constructs such that,
For "qwerty = 4," ruby UNAMBIGUOUSLY
invokes an existing setter if there
is no local variable defined
You say no. I say yes; another ruby could exist which behaves exactly like
the current in every respect, except "qwerty = 4" defines a new
variable when no setter and no local exists, it invokes the setter if one
exists, and it assigns to the local if one exists. I fully accept that I
could be wrong. In fact, a reason why I might be wrong would be interesting.
Let me explain.
Imagine you are writing a new OO language with accessor methods looking
like instances vars (like ruby & C#). You'd probably start with
conceptual grammars something like:
var = expr // assignment
method = expr // setter method invocation
But the parser-compiler (not even the runtime) will puke, because even after
all the input is grokked there's no way to know which grammar is pertinent.
You're faced which a classic choice. I can't be sure of the details, but
basically ruby does this:
var = expr // assignment (new or existing)
// method = expr, disallow setter method invocation without .
that is why it's un-Ambiguous, while and C# does this:
symbol = expr // push 'symbol=' onto parse tree and decide later
// if local variable is def'd somewhere in scope: assignment
// else if a setter is def'd in scope: invocation
For C#, 'later' is still at compile time.
I'm sure ruby could do the same, but 'later' would have to be at runtime, because
as ben points out you don't know until the statement is executed which case
applies.
My question was never intended to mean "do I really need the 'self.'?" or "what
potential ambiguity is being avoided?" Rather I wanted to know why was this
particular choice made? Maybe it's not performance. Maybe it just got the job
done, or it was considered best to always allow a 1-liner local to override a
method (a pretty rare case requirement) ...
But I'm sort of suggesting that the most dynamical language might be the one which
postpones this decision the longest, and chooses semantics based on the most contextual
info: so if you have no local and you defined a setter, it would use the setter. Isn't
this why we like ruby, smalltalk, objc, because method invocation is decided at runtime,
offering maximum expressiveness?
Well, I think the reason this is the case is because qwerty = 4 is ambiguous—are you defining a new variable called qwerty or calling the setter? Ruby resolves this ambiguity by saying it will create a new variable, thus the self. is required.
Here is another case where you need self.:
class A
def test
4
end
def use_variable
test = 5
test
end
def use_method
test = 5
self.test
end
end
a = A.new
a.use_variable # returns 5
a.use_method # returns 4
As you can see, the access to test is ambiguous, so the self. is required.
Also, this is why the C# example is actually not a good comparison, because you define variables in a way that is unambiguous from using the setter. If you had defined a variable in C# that was the same name as the accessor, you would need to qualify calls to the accessor with this., just like the Ruby case.
The important thing to remember here is that Ruby methods can be (un)defined at any point, so to intelligently resolve the ambiguity, every assignment would need to run code to check whether there is a method with the assigned-to name at the time of assignment.
Because otherwise it would be impossible to set local variables at all inside of methods. variable = some_value is ambiguous. For example:
class ExampleClass
attr_reader :last_set
def method_missing(name, *args)
if name.to_s =~ /=$/
#last_set = args.first
else
super
end
end
def some_method
some_variable = 5 # Set a local variable? Or call method_missing?
puts some_variable
end
end
If self wasn't required for setters, some_method would raise NameError: undefined local variable or method 'some_variable'. As-is though, the method works as intended:
example = ExampleClass.new
example.blah = 'Some text'
example.last_set #=> "Some text"
example.some_method # prints "5"
example.last_set #=> "Some text"

closure verification override default mocking

I have a method looks like this:
public void save(DbSession session,Wrappe wrapper,Wrappe wrappe){
//...other logic
//save wrapper
wrapper=(Wrapper)session.save(wrapper)
//set wrapper's id into wrappee
wrappee.setWrapperId(wrapper.getId());
//save wrappee
session.save(wrappee);
}
and test code looks like this:
given:
session.save(_) >> wrapperWithGeneratedId
when:
obj.save(session,wrapper,wrappee)
then:"wrapper got saved"
1*session.save(_) >> {Wrapper save ->
diffs(wrapper,saved)==null
}
and:"wrappee"
1*session.save(_) >> {Wrappe saved ->
diffs(wrappee,saved)==null
}
These test code will give an exception:
java.lang.ClassCastException: java.lang.Boolean cannot be cast to com.company.model.Wrapper
If commented verification closure in "then" section,test will pass,so I guess this section
1*session.save(_) >> {Wrapper save ->
diffs(wrapper,saved)==null
}
overrode this mocking:
session.save(_) >> wrapperWithGeneratedId
Is any way do both correctly?
1st. 'and' is syntactic sugar. It's just a way to visually separate code within the same block. Your last two mocks are effectively the same (although since you're testing behaviorally it will still verify that save is called twice.)
2nd. Assuming you want to verify thatdiffs(wrapper,saved)==null, that won't currently happen because it's not a 'base level' evaluation. Anything within then/where/closures/etc needs to be prepended with 'assert ' if you want to evaluate it.
3rd. A then block is scoped to its when block and can override existing mocks; your assumption that your mock was being overwritten is correct.
4th. Is there any reason you don't just include your return value alongside your evaluation?
2 * session.save(_) >> {Wrapper save ->
diffs(wrapper,saved)==null
return wrapperWithGeneratedId
}
5th. Your error is due to your mock returning a boolean (your assertion logic) which Groovy then tries (and fails) to parse into a Wrapper. My assumption for why this is happening is that .save() has a return type of Wrapper. To fix that you will either need to create a boolean constructor for Wrapper, or change your mock to return something Groovy can turn into a Wrapper (how-to in point 4)
Official Stub/Mock/Spy documentation (quite good, worth a read)

Grails integration test - domain object equality

Setting up some integration tests, I'm having issues with domain class equality. The equality works as expected during normal execution, but when testing the Service methods through an integration test, the test for equality is coming back false.
One service (called in the setUp() of the Test Case) puts a Domain object into the session
SomeService {
setSessionVehicle(String name) {
Vehicle vehicle = Vehicle.findByName(name)
session.setAttribute("SessionVehicle", vehicle)
}
getSessionVehicle() {
return session.getAttribute("SessionVehicle")
}
}
Elsewhere in another service, I load an object and make sure the associated attribute object matches the session value:
OtherService {
getEngine(long id) {
Vehicle current = session.getAttribute("SessionVehicle")
Engine engine = Engine.get(id)
if(!engine.vehicle.equals(current)) throw Exception("Blah blah")
}
}
This works as expected during normal operation, preventing from loading the wrong engine (Ok, I sanitized the class names, pretend it makes sense). But in the integration test, that .equals() fails when it should succeed:
Vehicle testVehicle
setUp() {
Vehicle v = new Vehicle("Default")
v.save()
someService.setSessionVehicle("Default")
testVehicle = someService.getSessionVehicle()
}
testGetEngine() {
List<Engine> engines = Engine.findAllByVehicle(testVehicle)
//some assertions and checks
Engine e = otherService.getEngine(engines.get(0).id)
}
The findAll() call is correctly returning the list of all Engines associated with the vehicle in the session, but when I try to look up an individual Engine by ID, the equality check for session Vehicle vs Vehicle on the found Engine fails. Only a single vehicle has been created at this point and the Exception message displays that the session Vehicle and the Engine.Vehicle exist and are the same value.
If I attempt this equality check in the testCase itself, it fails, but I'm able to change the testCase to check if(vehicle.id == sessionVehicle.id) which succeeds, but I'm not keen on changing my production code in order to satisfy an integration test.
Am I doing something wrong when setting up these domain objects in my test case that I should be doing differently?
First of all, the equality check you are doing is just checking the reference. You should not use the default equals method for your check, better override the equals method in domain class.
There are two ways you can override the equals method:
1) you can use your IDE to auto-generate code for equals method(a lot of null checking etc..).
2) Preferred way: You can use EqualsBuilder and HashCodeBuilder classes from the Apache Commons project. The library should be already available to your application, or download the JAR file and place in lib. Here is sample code for using EqualsBuilder:
boolean equals(o) {
if ( !(o instanceof Vehicle) ) {
return false
}
def eb = new EqualsBuilder()
eb.append(id, o.id)
eb.append(name, o.name)
eb.append(otherProperties, o.otherProperties)
....
return eb.isEquals()
}
Another point is: how you are getting session in service? from RequestContextHolder? Its a good practice to not access session directly from service, rather send the value as method parameter in the service.

Unexpected Validate behavior with Moq

Moq has been driving me a bit crazy on my latest project. I recently upgraded to version 4.0.10827, and I'm noticing what seems to me to be a new behavior.
Basically, when I call my mocked function (MakeCall, in this example) in the code I am testing, I am passing in an object (TestClass). The code I am testing makes changes to the TestClass object before and after the call to MakeCall. Once the code has completed, I then call Moq's Verify function. My expectation is that Moq will have recorded the complete object that I passed into MakeCall, perhaps via a mechanism like deep cloning. This way, I will be able to verify that MakeCall was called with the exact object I am expecting it to be called with. Unfortunately, this is not what I'm seeing.
I attempt to illustrate this in the code below (hopefully, clarifying it a bit in the process).
I first create a new TestClass object. Its Var property is set to "one".
I then create the mocked object, mockedObject, which is my test subject.
I then call the MakeCall method of mockedObject (by the way, the Machine.Specifications framework used in the example allows the code in the When_Testing class to be read from top to bottom).
I then test the mocked object to ensure that it was indeed called with a TestClass with a Var value of "one". This succeeds, as I expected it to.
I then make a change to the original TestClass object by re-assigning the Var property to "two".
I then proceed to attempt to verify if Moq still thinks that MakeCall was called with a TestClass with a value of "one". This fails, although I am expecting it to be true.
Finally, I test to see if Moq thinks MakeCall was in fact called by a TestClass object with a value of "two". This succeeds, although I would initially have expected it to fail.
It seems pretty clear to me that Moq is only holding onto a reference to the original TestClass object, allowing me to change its value with impunity, adversely affecting the results of my testing.
A few notes on the test code. IMyMockedInterface is the interface I am mocking. TestClass is the class I am passing into the MakeCall method and therefore using to demonstrate the issue I am having. Finally, When_Testing is the actual test class that contains the test code. It is using the Machine.Specifications framework, which is why there are a few odd items ('Because of', 'It should...'). These are simply delegates that are called by the framework to execute the tests. They should be easily removed and the contained code placed into a standard function if that is desired. I left it in this format because it allows all Validate calls to complete (as compared to the 'Arrange, Act Assert' paradigm). Just to clarify, the below code is not the actual code I am having problems with. It is simply intended to illustrate the problem, as I have seen this same behavior in multiple places.
using Machine.Specifications;
// Moq has a conflict with MSpec as they both have an 'It' object.
using moq = Moq;
public interface IMyMockedInterface
{
int MakeCall(TestClass obj);
}
public class TestClass
{
public string Var { get; set; }
// Must override Equals so Moq treats two objects with the
// same value as equal (instead of comparing references).
public override bool Equals(object obj)
{
if ((obj != null) && (obj.GetType() != this.GetType()))
return false;
TestClass t = obj as TestClass;
if (t.Var != this.Var)
return false;
return true;
}
public override int GetHashCode()
{
int hash = 41;
int factor = 23;
hash = (hash ^ factor) * Var.GetHashCode();
return hash;
}
public override string ToString()
{
return MvcTemplateApp.Utilities.ClassEnhancementUtilities.ObjectToString(this);
}
}
[Subject(typeof(object))]
public class When_Testing
{
// TestClass is set up to contain a value of 'one'
protected static TestClass t = new TestClass() { Var = "one" };
protected static moq.Mock<IMyMockedInterface> mockedObject = new moq.Mock<IMyMockedInterface>();
Because of = () =>
{
mockedObject.Object.MakeCall(t);
};
// Test One
// Expected: Moq should verify that MakeCall was called with a TestClass with a value of 'one'.
// Actual: Moq does verify that MakeCall was called with a TestClass with a value of 'one'.
// Result: This is correct.
It should_verify_that_make_call_was_called_with_a_value_of_one = () =>
mockedObject.Verify(o => o.MakeCall(new TestClass() { Var = "one" }), moq.Times.Once());
// Update the original object to contain a new value.
It should_update_the_test_class_value_to_two = () =>
t.Var = "two";
// Test Two
// Expected: Moq should verify that MakeCall was called with a TestClass with a value of 'one'.
// Actual: The Verify call fails, claiming that MakeCall was never called with a TestClass instance with a value of 'one'.
// Result: This is incorrect.
It should_verify_that_make_call_was_called_with_a_class_containing_a_value_of_one = () =>
mockedObject.Verify(o => o.MakeCall(new TestClass() { Var = "one" }), moq.Times.Once());
// Test Three
// Expected: Moq should fail to verify that MakeCall was called with a TestClass with a value of 'two'.
// Actual: Moq actually does verify that MakeCall was called with a TestClass with a value of 'two'.
// Result: This is incorrect.
It should_fail_to_verify_that_make_call_was_called_with_a_class_containing_a_value_of_two = () =>
mockedObject.Verify(o => o.MakeCall(new TestClass() { Var = "two" }), moq.Times.Once());
}
I have a few questions regarding this:
Is this expected behavior?
Is this new behavior?
Is there a workaround that I am unaware of?
Am I using Verify incorrectly?
Is there a better way of using Moq to avoid this situation?
I thank you humbly for any assistance you can provide.
Edit:
Here is one of the actual tests and SUT code that I experienced this problem with. Hopefully it will act as clarification.
// This is the MVC Controller Action that I am testing. Note that it
// makes changes to the 'searchProjects' object before and after
// calling 'repository.SearchProjects'.
[HttpGet]
public ActionResult List(int? page, [Bind(Include = "Page, SearchType, SearchText, BeginDate, EndDate")]
SearchProjects searchProjects)
{
int itemCount;
searchProjects.ItemsPerPage = profile.ItemsPerPage;
searchProjects.Projects = repository.SearchProjects(searchProjects,
profile.UserKey, out itemCount);
searchProjects.TotalItems = itemCount;
return View(searchProjects);
}
// This is my test class for the controller's List action. The controller
// is instantiated in an Establish delegate in the 'with_project_controller'
// class, along with the SearchProjectsRequest, SearchProjectsRepositoryGet,
// and SearchProjectsResultGet objects which are defined below.
[Subject(typeof(ProjectController))]
public class When_the_project_list_method_is_called_via_a_get_request
: with_project_controller
{
protected static int itemCount;
protected static ViewResult result;
Because of = () =>
result = controller.List(s.Page, s.SearchProjectsRequest) as ViewResult;
// This test fails, as it is expecting the 'SearchProjects' object
// to contain:
// Page, SearchType, SearchText, BeginDate, EndDate and ItemsPerPage
It should_call_the_search_projects_repository_method = () =>
s.Repository.Verify(r => r.SearchProjects(s.SearchProjectsRepositoryGet,
s.UserKey, out itemCount), moq.Times.Once());
// This test succeeds, as it is expecting the 'SearchProjects' object
// to contain:
// Page, SearchType, SearchText, BeginDate, EndDate, ItemsPerPage,
// Projects and TotalItems
It should_call_the_search_projects_repository_method = () =>
s.Repository.Verify(r => r.SearchProjects(s.SearchProjectsResultGet,
s.UserKey, out itemCount), moq.Times.Once());
It should_return_the_correct_view_name = () =>
result.ViewName.ShouldBeEmpty();
It should_return_the_correct_view_model = () =>
result.Model.ShouldEqual(s.SearchProjectsResultGet);
}
/////////////////////////////////////////////////////
// Here are the values of the three test objects
/////////////////////////////////////////////////////
// This is the object that is returned by the client.
SearchProjects SearchProjectsRequest = new SearchProjects()
{
SearchType = SearchTypes.ProjectName,
SearchText = GetProjectRequest().Name,
Page = Page
};
// This is the object I am expecting the repository method to be called with.
SearchProjects SearchProjectsRepositoryGet = new SearchProjects()
{
SearchType = SearchTypes.ProjectName,
SearchText = GetProjectRequest().Name,
Page = Page,
ItemsPerPage = ItemsPerPage
};
// This is the complete object I expect to be returned to the view.
SearchProjects SearchProjectsResultGet = new SearchProjects()
{
SearchType = SearchTypes.ProjectName,
SearchText = GetProjectRequest().Name,
Page = Page,
ItemsPerPage = ItemsPerPage,
Projects = new List<Project>() { GetProjectRequest() },
TotalItems = TotalItems
};
Ultimately, your question is whether a mocking framework should take snapshots of the parameters you use when interacting with the mocks so that it can accurately record the state the system was in at the point of interaction rather than the state the parameters might be in at the point of verification.
I would say this is a reasonable expectation from a logical point of view. You are performing action X with value Y. If you ask the mock "Did I perform action X with value Y", you expect it to say "Yes" regardless of the current state of the system.
To summarize the problem you are running into:
You first invoke a method on a mock object with a reference type parameter.
Moq saves information about the invocation along with the reference type parameter passed in.
You then ask Moq if the method was called one time with an object equal to the reference you passed in.
Moq checks its history for a call to that method with a parameter that matches the supplied parameter and answers yes.
You then modify the object that you passed as the parameter to the method call on the mock.
The memory space of the reference Moq is holding in its history changes to the new value.
You then ask Moq if the method was called one time with an object that isn't equal to the reference its holding.
Mock checks its history for a call to that method with a parameter that matches the supplied parameter and reports no.
To attempt to answer your specific questions:
Is this expected behavior?
I would say no.
Is this new behavior?
I don't know, but it's doubtful the project would have at one time had behavior that facilitated this and was later modified to only allow the simple scenario of only verifying a single usage per mock.
Is there a workaround that I am unaware of?
I'll answer this two ways.
From a technical standpoint, a workaround would be to use a Test Spy rather than a Mock. By using a Test Spy, you can record the values passed and use your own strategy for remembering the state, such as doing a deep clone, serializing the object, or just storing the specific values you care about to be compared against later.
From a testing standpoint, I would recommend that you follow the principle "Use The Front Door First". I believe there is a time for state-based testing as well as interaction-based testing, but you should try to avoid coupling yourself to the implementation details unless the interaction is an important part of the scenario. In some cases, the scenario you are interested in will be primarily about interaction ("Transfer funds between accounts"), but in other cases all you really care about is getting the correct result ("Withdraw $10"). In the case of the specification for your controller, this seems to fall into the query category, not the command category. You don't really care how it gets the results you want as long as they are correct. Therefore, I would recommend using state-based testing in this case. If another specification concerns issuing a command against the system, there may still end up being a front door solution which you should consider using first, but it may be necessary or important to do interaction based testing. Just my thoughts though.
Am I using Verify incorrectly?
You are using the Verify() method correctly, it just doesn't support the scenario you are using it for.
Is there a better way of using Moq to avoid this situation?
I don't think Moq is currently implemented to handle this scenario.
Hope this helps,
Derek Greer
http://derekgreer.lostechies.com
http://aspiringcraftsman.com
#derekgreer
First, you can avoid the conflict between Moq and MSpec by declaring
using Machine.Specifications;
using Moq;
using It = Machine.Specifications.It;
Then you'll only need to prefix with Moq. when you want to use Moq's It, for example Moq.It.IsAny<>().
Onto your question.
Note: This is not the original answer but an edited one after the OP added some real example code to the question
I've been trying out your sample code and I think it's got more to do with MSpec than Moq. Apparently (and I didn't know this either), when you modify the state of your SUT (System Under Test) inside an It delegate the changes gets remembered. What is happening now is:
Because delegate is run
It delegates are run, one after the other. If one changes the state, the following It will never see the set up in the Because. Hence your failed test.
I've tried marking your spec with the SetupForEachSpecificationAttribute:
[Subject(typeof(object)), SetupForEachSpecification]
public class When_Testing
{
// Something, Something, something...
}
The attribute does as its name says: It will run your Establish and Because before every It. Adding the attribute made the spec behave as expected: 3 Successes, one fail (the verification that with Var = "two").
Would the SetupForEachSpecificationAttribute solve your problem or is resetting after every It not acceptable for your tests?
FYI: I'm using Moq v4.0.10827.0 and MSpec v0.4.9.0
Free tip #2: If you're testing ASP.NET MVC apps with Mspec you might want to take a look at James Broome's MSpec extensions for MVC

Resources