This is my first experience with downcasting in C++ and I just can't understand the problem.
AInstruction and CInstruction inherit from AssemblerInstruction.
Parser takes the info in its ctor and creates one of those derived instruction types for its mInstruction member (accessed by getInstruction). In the program, a method of the base AssemblerInstruction class is used, for happy polymorphism.
But when I want to test that the Parser has created the correct instruction, I need to query the derived instruction members, which means I need to downcast parser.getInstruction() to an AInstruction or CInstruction.
As far as I can tell this needs to be done using a bunch of pointers and references. This is how I can get the code to compile:
TEST(ParserA, parsesBuiltInConstants)
{
AssemblerInstruction inst = Parser("#R3", 0).getInstruction();
EXPECT_EQ(inst.getInstructionType(), AssemblerInstruction::InstructionType::A);
AssemblerInstruction* i = &(inst);
AInstruction* a = dynamic_cast<AInstruction*>(i);
EXPECT_EQ(a->getLine(), "R3");
}
Running this gives this error:
unknown file: error: SEH exception with code 0xc0000005 thrown in the test body.
And stepping through the code, when the debugger is on the final line of the function, a is pointing to
0x00000000 <NULL>.
I imagine this is an instance where I don't have a full enough understanding of C++, meaning that I could be making a n00b mistake. Or maybe it's some bigger crazy problem. Help?
Update
I've been able to make this work by making mInstruction into a (dumb) pointer:
// in parser, when parsing
mInstructionPtr = new AInstruction(assemblyCode.substr(1), lineNumber);
// elsewhere in AssemblerInstruction.cpp
AssemblerInstruction* AssemblyParser::getInstructionPtr() { return mInstructionPtr; }
TEST(ParserA, parsesBuiltInConstants)
{
auto ptr = Parser("#R3", 0).getInstructionPtr();
AInstruction* a = dynamic_cast<AInstruction*>(ptr);
EXPECT_EQ(a->getLine(), "R3");
}
However I have trouble implementing it with a unique_ptr:
(I'm aware that mInstruction (non-pointer) is redundant, as are two types of pointers. I'll get rid of it later when I clean all this up)
class AssemblyParser
{
public:
AssemblyParser(std::string assemblyCode, unsigned int lineNumber);
AssemblerInstruction getInstruction();
std::unique_ptr<AssemblerInstruction> getUniqueInstructionPtr();
AssemblerInstruction* getInstructionPtr();
private:
AssemblerInstruction mInstruction;
std::unique_ptr<AssemblerInstruction> mUniqueInstructionPtr;
AssemblerInstruction* mInstructionPtr;
};
// in AssemblyParser.cpp
// in parser as in example above. this works fine.
mUniqueInstructionPtr = make_unique<AInstruction>(assemblyCode.substr(1), lineNumber);
// this doesn't compile!!!
unique_ptr<AssemblerInstruction> AssemblyParser::getUniqueInstructionPtr()
{
return mUniqueInstructionPtr;
}
In getUniqueInstructionPtr, there is a squiggle under mUniqueInstructionPtr with this error:
'std::unique_ptr<AssemblerInstruction,std::default_delete>::unique_ptr(const std::unique_ptr<AssemblerInstruction,std::default_delete> &)': attempting to reference a deleted function
What!? I haven't declared any functions as deleted or defaulted!
You can not downcast an object to something which doesn't match it's dynamic type. In your code,
AssemblerInstruction inst = Parser("#R3", 0).getInstruction();
inst has a fixed type, which is AssemblerInstruction. Downcasting it to AInstruction leads to undefined behavior - manifested as crash - because that is not what it is.
If you want your getInstruction to return a dynamically-typed object, it has to return a [smart] pointer to base class, while constructing an object of derived class. Something like that (pseudo code):
std::unique_ptr<AssemblerInstruction> getInstruction(...) {
return std::make_unique<AInstruction>(...);
}
Also, if you see yourself in need of downcasting object based on a value of a class, you are doing something wrong, as you are trying to home-brew polymorphism. Most of the times it does indicate a design flaw, and should instead be done using built-in C++ polymorphic support - namely, virtual functions.
This question's answers are a community effort. Edit existing answers to improve this post. It is not currently accepting new answers or interactions.
What are Null Pointer Exceptions (java.lang.NullPointerException) and what causes them?
What methods/tools can be used to determine the cause so that you stop the exception from causing the program to terminate prematurely?
There are two overarching types of variables in Java:
Primitives: variables that contain data. If you want to manipulate the data in a primitive variable you can manipulate that variable directly. By convention primitive types start with a lowercase letter. For example variables of type int or char are primitives.
References: variables that contain the memory address of an Object i.e. variables that refer to an Object. If you want to manipulate the Object that a reference variable refers to you must dereference it. Dereferencing usually entails using . to access a method or field, or using [ to index an array. By convention reference types are usually denoted with a type that starts in uppercase. For example variables of type Object are references.
Consider the following code where you declare a variable of primitive type int and don't initialize it:
int x;
int y = x + x;
These two lines will crash the program because no value is specified for x and we are trying to use x's value to specify y. All primitives have to be initialized to a usable value before they are manipulated.
Now here is where things get interesting. Reference variables can be set to null which means "I am referencing nothing". You can get a null value in a reference variable if you explicitly set it that way, or a reference variable is uninitialized and the compiler does not catch it (Java will automatically set the variable to null).
If a reference variable is set to null either explicitly by you or through Java automatically, and you attempt to dereference it you get a NullPointerException.
The NullPointerException (NPE) typically occurs when you declare a variable but did not create an object and assign it to the variable before trying to use the contents of the variable. So you have a reference to something that does not actually exist.
Take the following code:
Integer num;
num = new Integer(10);
The first line declares a variable named num, but it does not actually contain a reference value yet. Since you have not yet said what to point to, Java sets it to null.
In the second line, the new keyword is used to instantiate (or create) an object of type Integer, and the reference variable num is assigned to that Integer object.
If you attempt to dereference num before creating the object you get a NullPointerException. In the most trivial cases, the compiler will catch the problem and let you know that "num may not have been initialized," but sometimes you may write code that does not directly create the object.
For instance, you may have a method as follows:
public void doSomething(SomeObject obj) {
// Do something to obj, assumes obj is not null
obj.myMethod();
}
In which case, you are not creating the object obj, but rather assuming that it was created before the doSomething() method was called. Note, it is possible to call the method like this:
doSomething(null);
In which case, obj is null, and the statement obj.myMethod() will throw a NullPointerException.
If the method is intended to do something to the passed-in object as the above method does, it is appropriate to throw the NullPointerException because it's a programmer error and the programmer will need that information for debugging purposes.
In addition to NullPointerExceptions thrown as a result of the method's logic, you can also check the method arguments for null values and throw NPEs explicitly by adding something like the following near the beginning of a method:
// Throws an NPE with a custom error message if obj is null
Objects.requireNonNull(obj, "obj must not be null");
Note that it's helpful to say in your error message clearly which object cannot be null. The advantage of validating this is that 1) you can return your own clearer error messages and 2) for the rest of the method you know that unless obj is reassigned, it is not null and can be dereferenced safely.
Alternatively, there may be cases where the purpose of the method is not solely to operate on the passed in object, and therefore a null parameter may be acceptable. In this case, you would need to check for a null parameter and behave differently. You should also explain this in the documentation. For example, doSomething() could be written as:
/**
* #param obj An optional foo for ____. May be null, in which case
* the result will be ____.
*/
public void doSomething(SomeObject obj) {
if(obj == null) {
// Do something
} else {
// Do something else
}
}
Finally, How to pinpoint the exception & cause using Stack Trace
What methods/tools can be used to determine the cause so that you stop
the exception from causing the program to terminate prematurely?
Sonar with find bugs can detect NPE.
Can sonar catch null pointer exceptions caused by JVM Dynamically
Now Java 14 has added a new language feature to show the root cause of NullPointerException. This language feature has been part of SAP commercial JVM since 2006.
In Java 14, the following is a sample NullPointerException Exception message:
in thread "main" java.lang.NullPointerException: Cannot invoke "java.util.List.size()" because "list" is null
List of situations that cause a NullPointerException to occur
Here are all the situations in which a NullPointerException occurs, that are directly* mentioned by the Java Language Specification:
Accessing (i.e. getting or setting) an instance field of a null reference. (static fields don't count!)
Calling an instance method of a null reference. (static methods don't count!)
throw null;
Accessing elements of a null array.
Synchronising on null - synchronized (someNullReference) { ... }
Any integer/floating point operator can throw a NullPointerException if one of its operands is a boxed null reference
An unboxing conversion throws a NullPointerException if the boxed value is null.
Calling super on a null reference throws a NullPointerException. If you are confused, this is talking about qualified superclass constructor invocations:
class Outer {
class Inner {}
}
class ChildOfInner extends Outer.Inner {
ChildOfInner(Outer o) {
o.super(); // if o is null, NPE gets thrown
}
}
Using a for (element : iterable) loop to loop through a null collection/array.
switch (foo) { ... } (whether its an expression or statement) can throw a NullPointerException when foo is null.
foo.new SomeInnerClass() throws a NullPointerException when foo is null.
Method references of the form name1::name2 or primaryExpression::name throws a NullPointerException when evaluated when name1 or primaryExpression evaluates to null.
a note from the JLS here says that, someInstance.someStaticMethod() doesn't throw an NPE, because someStaticMethod is static, but someInstance::someStaticMethod still throw an NPE!
* Note that the JLS probably also says a lot about NPEs indirectly.
NullPointerExceptions are exceptions that occur when you try to use a reference that points to no location in memory (null) as though it were referencing an object. Calling a method on a null reference or trying to access a field of a null reference will trigger a NullPointerException. These are the most common, but other ways are listed on the NullPointerException javadoc page.
Probably the quickest example code I could come up with to illustrate a NullPointerException would be:
public class Example {
public static void main(String[] args) {
Object obj = null;
obj.hashCode();
}
}
On the first line inside main, I'm explicitly setting the Object reference obj equal to null. This means I have a reference, but it isn't pointing to any object. After that, I try to treat the reference as though it points to an object by calling a method on it. This results in a NullPointerException because there is no code to execute in the location that the reference is pointing.
(This is a technicality, but I think it bears mentioning: A reference that points to null isn't the same as a C pointer that points to an invalid memory location. A null pointer is literally not pointing anywhere, which is subtly different than pointing to a location that happens to be invalid.)
What is a NullPointerException?
A good place to start is the JavaDocs. They have this covered:
Thrown when an application attempts to use null in a case where an
object is required. These include:
Calling the instance method of a null object.
Accessing or modifying the field of a null object.
Taking the length of null as if it were an array.
Accessing or modifying the slots of null as if it were an array.
Throwing null as if it were a Throwable value.
Applications should throw instances of this class to indicate other
illegal uses of the null object.
It is also the case that if you attempt to use a null reference with synchronized, that will also throw this exception, per the JLS:
SynchronizedStatement:
synchronized ( Expression ) Block
Otherwise, if the value of the Expression is null, a NullPointerException is thrown.
How do I fix it?
So you have a NullPointerException. How do you fix it? Let's take a simple example which throws a NullPointerException:
public class Printer {
private String name;
public void setName(String name) {
this.name = name;
}
public void print() {
printString(name);
}
private void printString(String s) {
System.out.println(s + " (" + s.length() + ")");
}
public static void main(String[] args) {
Printer printer = new Printer();
printer.print();
}
}
Identify the null values
The first step is identifying exactly which values are causing the exception. For this, we need to do some debugging. It's important to learn to read a stacktrace. This will show you where the exception was thrown:
Exception in thread "main" java.lang.NullPointerException
at Printer.printString(Printer.java:13)
at Printer.print(Printer.java:9)
at Printer.main(Printer.java:19)
Here, we see that the exception is thrown on line 13 (in the printString method). Look at the line and check which values are null by
adding logging statements or using a debugger. We find out that s is null, and calling the length method on it throws the exception. We can see that the program stops throwing the exception when s.length() is removed from the method.
Trace where these values come from
Next check where this value comes from. By following the callers of the method, we see that s is passed in with printString(name) in the print() method, and this.name is null.
Trace where these values should be set
Where is this.name set? In the setName(String) method. With some more debugging, we can see that this method isn't called at all. If the method was called, make sure to check the order that these methods are called, and the set method isn't called after the print method.
This is enough to give us a solution: add a call to printer.setName() before calling printer.print().
Other fixes
The variable can have a default value (and setName can prevent it being set to null):
private String name = "";
Either the print or printString method can check for null, for example:
printString((name == null) ? "" : name);
Or you can design the class so that name always has a non-null value:
public class Printer {
private final String name;
public Printer(String name) {
this.name = Objects.requireNonNull(name);
}
public void print() {
printString(name);
}
private void printString(String s) {
System.out.println(s + " (" + s.length() + ")");
}
public static void main(String[] args) {
Printer printer = new Printer("123");
printer.print();
}
}
See also:
Avoiding “!= null” statements in Java?
I still can't find the problem
If you tried to debug the problem and still don't have a solution, you can post a question for more help, but make sure to include what you've tried so far. At a minimum, include the stacktrace in the question, and mark the important line numbers in the code. Also, try simplifying the code first (see SSCCE).
Question: What causes a NullPointerException (NPE)?
As you should know, Java types are divided into primitive types (boolean, int, etc.) and reference types. Reference types in Java allow you to use the special value null which is the Java way of saying "no object".
A NullPointerException is thrown at runtime whenever your program attempts to use a null as if it was a real reference. For example, if you write this:
public class Test {
public static void main(String[] args) {
String foo = null;
int length = foo.length(); // HERE
}
}
the statement labeled "HERE" is going to attempt to run the length() method on a null reference, and this will throw a NullPointerException.
There are many ways that you could use a null value that will result in a NullPointerException. In fact, the only things that you can do with a null without causing an NPE are:
assign it to a reference variable or read it from a reference variable,
assign it to an array element or read it from an array element (provided that array reference itself is non-null!),
pass it as a parameter or return it as a result, or
test it using the == or != operators, or instanceof.
Question: How do I read the NPE stacktrace?
Suppose that I compile and run the program above:
$ javac Test.java
$ java Test
Exception in thread "main" java.lang.NullPointerException
at Test.main(Test.java:4)
$
First observation: the compilation succeeds! The problem in the program is NOT a compilation error. It is a runtime error. (Some IDEs may warn your program will always throw an exception ... but the standard javac compiler doesn't.)
Second observation: when I run the program, it outputs two lines of "gobbledy-gook". WRONG!! That's not gobbledy-gook. It is a stacktrace ... and it provides vital information that will help you track down the error in your code if you take the time to read it carefully.
So let's look at what it says:
Exception in thread "main" java.lang.NullPointerException
The first line of the stack trace tells you a number of things:
It tells you the name of the Java thread in which the exception was thrown. For a simple program with one thread (like this one), it will be "main". Let's move on ...
It tells you the full name of the exception that was thrown; i.e. java.lang.NullPointerException.
If the exception has an associated error message, that will be output after the exception name. NullPointerException is unusual in this respect, because it rarely has an error message.
The second line is the most important one in diagnosing an NPE.
at Test.main(Test.java:4)
This tells us a number of things:
"at Test.main" says that we were in the main method of the Test class.
"Test.java:4" gives the source filename of the class, AND it tells us that the statement where this occurred is in line 4 of the file.
If you count the lines in the file above, line 4 is the one that I labeled with the "HERE" comment.
Note that in a more complicated example, there will be lots of lines in the NPE stack trace. But you can be sure that the second line (the first "at" line) will tell you where the NPE was thrown1.
In short, the stack trace will tell us unambiguously which statement of the program has thrown the NPE.
See also: What is a stack trace, and how can I use it to debug my application errors?
1 - Not quite true. There are things called nested exceptions...
Question: How do I track down the cause of the NPE exception in my code?
This is the hard part. The short answer is to apply logical inference to the evidence provided by the stack trace, the source code, and the relevant API documentation.
Let's illustrate with the simple example (above) first. We start by looking at the line that the stack trace has told us is where the NPE happened:
int length = foo.length(); // HERE
How can that throw an NPE?
In fact, there is only one way: it can only happen if foo has the value null. We then try to run the length() method on null and... BANG!
But (I hear you say) what if the NPE was thrown inside the length() method call?
Well, if that happened, the stack trace would look different. The first "at" line would say that the exception was thrown in some line in the java.lang.String class and line 4 of Test.java would be the second "at" line.
So where did that null come from? In this case, it is obvious, and it is obvious what we need to do to fix it. (Assign a non-null value to foo.)
OK, so let's try a slightly more tricky example. This will require some logical deduction.
public class Test {
private static String[] foo = new String[2];
private static int test(String[] bar, int pos) {
return bar[pos].length();
}
public static void main(String[] args) {
int length = test(foo, 1);
}
}
$ javac Test.java
$ java Test
Exception in thread "main" java.lang.NullPointerException
at Test.test(Test.java:6)
at Test.main(Test.java:10)
$
So now we have two "at" lines. The first one is for this line:
return args[pos].length();
and the second one is for this line:
int length = test(foo, 1);
Looking at the first line, how could that throw an NPE? There are two ways:
If the value of bar is null then bar[pos] will throw an NPE.
If the value of bar[pos] is null then calling length() on it will throw an NPE.
Next, we need to figure out which of those scenarios explains what is actually happening. We will start by exploring the first one:
Where does bar come from? It is a parameter to the test method call, and if we look at how test was called, we can see that it comes from the foo static variable. In addition, we can see clearly that we initialized foo to a non-null value. That is sufficient to tentatively dismiss this explanation. (In theory, something else could change foo to null ... but that is not happening here.)
So what about our second scenario? Well, we can see that pos is 1, so that means that foo[1] must be null. Is this possible?
Indeed it is! And that is the problem. When we initialize like this:
private static String[] foo = new String[2];
we allocate a String[] with two elements that are initialized to null. After that, we have not changed the contents of foo ... so foo[1] will still be null.
What about on Android?
On Android, tracking down the immediate cause of an NPE is a bit simpler. The exception message will typically tell you the (compile time) type of the null reference you are using and the method you were attempting to call when the NPE was thrown. This simplifies the process of pinpointing the immediate cause.
But on the flipside, Android has some common platform-specific causes for NPEs. A very common is when getViewById unexpectedly returns a null. My advice would be to search for Q&As about the cause of the unexpected null return value.
It's like you are trying to access an object which is null. Consider below example:
TypeA objA;
At this time you have just declared this object but not initialized or instantiated. And whenever you try to access any property or method in it, it will throw NullPointerException which makes sense.
See this below example as well:
String a = null;
System.out.println(a.toString()); // NullPointerException will be thrown
A null pointer exception is thrown when an application attempts to use null in a case where an object is required. These include:
Calling the instance method of a null object.
Accessing or modifying the field of a null object.
Taking the length of null as if it were an array.
Accessing or modifying the slots of null as if it were an array.
Throwing null as if it were a Throwable value.
Applications should throw instances of this class to indicate other illegal uses of the null object.
Reference: http://docs.oracle.com/javase/8/docs/api/java/lang/NullPointerException.html
A null pointer is one that points to nowhere. When you dereference a pointer p, you say "give me the data at the location stored in "p". When p is a null pointer, the location stored in p is nowhere, you're saying "give me the data at the location 'nowhere'". Obviously, it can't do this, so it throws a null pointer exception.
In general, it's because something hasn't been initialized properly.
A lot of explanations are already present to explain how it happens and how to fix it, but you should also follow best practices to avoid NullPointerExceptions at all.
See also:
A good list of best practices
I would add, very important, make a good use of the final modifier.
Using the "final" modifier whenever applicable in Java
Summary:
Use the final modifier to enforce good initialization.
Avoid returning null in methods, for example returning empty collections when applicable.
Use annotations #NotNull and #Nullable
Fail fast and use asserts to avoid propagation of null objects through the whole application when they shouldn't be null.
Use equals with a known object first: if("knownObject".equals(unknownObject)
Prefer valueOf() over toString().
Use null safe StringUtils methods StringUtils.isEmpty(null).
Use Java 8 Optional as return value in methods, Optional class provide a solution for representing optional values instead of null references.
A null pointer exception is an indicator that you are using an object without initializing it.
For example, below is a student class which will use it in our code.
public class Student {
private int id;
public int getId() {
return this.id;
}
public setId(int newId) {
this.id = newId;
}
}
The below code gives you a null pointer exception.
public class School {
Student student;
public School() {
try {
student.getId();
}
catch(Exception e) {
System.out.println("Null pointer exception");
}
}
}
Because you are using student, but you forgot to initialize it like in the
correct code shown below:
public class School {
Student student;
public School() {
try {
student = new Student();
student.setId(12);
student.getId();
}
catch(Exception e) {
System.out.println("Null pointer exception");
}
}
}
In Java, everything (excluding primitive types) is in the form of a class.
If you want to use any object then you have two phases:
Declare
Initialization
Example:
Declaration: Object object;
Initialization: object = new Object();
Same for the array concept:
Declaration: Item item[] = new Item[5];
Initialization: item[0] = new Item();
If you are not giving the initialization section then the NullPointerException arise.
In Java all the variables you declare are actually "references" to the objects (or primitives) and not the objects themselves.
When you attempt to execute one object method, the reference asks the living object to execute that method. But if the reference is referencing NULL (nothing, zero, void, nada) then there is no way the method gets executed. Then the runtime let you know this by throwing a NullPointerException.
Your reference is "pointing" to null, thus "Null -> Pointer".
The object lives in the VM memory space and the only way to access it is using this references. Take this example:
public class Some {
private int id;
public int getId(){
return this.id;
}
public setId( int newId ) {
this.id = newId;
}
}
And on another place in your code:
Some reference = new Some(); // Point to a new object of type Some()
Some otherReference = null; // Initiallly this points to NULL
reference.setId( 1 ); // Execute setId method, now private var id is 1
System.out.println( reference.getId() ); // Prints 1 to the console
otherReference = reference // Now they both point to the only object.
reference = null; // "reference" now point to null.
// But "otherReference" still point to the "real" object so this print 1 too...
System.out.println( otherReference.getId() );
// Guess what will happen
System.out.println( reference.getId() ); // :S Throws NullPointerException because "reference" is pointing to NULL remember...
This an important thing to know - when there are no more references to an object (in the example above when reference and otherReference both point to null) then the object is "unreachable". There is no way we can work with it, so this object is ready to be garbage collected, and at some point, the VM will free the memory used by this object and will allocate another.
Another occurrence of a NullPointerException occurs when one declares an object array, then immediately tries to dereference elements inside of it.
String[] phrases = new String[10];
String keyPhrase = "Bird";
for(String phrase : phrases) {
System.out.println(phrase.equals(keyPhrase));
}
This particular NPE can be avoided if the comparison order is reversed; namely, use .equals on a guaranteed non-null object.
All elements inside of an array are initialized to their common initial value; for any type of object array, that means that all elements are null.
You must initialize the elements in the array before accessing or dereferencing them.
String[] phrases = new String[] {"The bird", "A bird", "My bird", "Bird"};
String keyPhrase = "Bird";
for(String phrase : phrases) {
System.out.println(phrase.equals(keyPhrase));
}
I want to create an error object. But there is no v8::Error::New() How can I create an error object?
v8::Handle< v8::Value > result = v8::Undefined();
v8::Handle< v8::Value > error = v8::Undefined();
if(m_errorMsg.empty())
{
// Not error
}
else
{
// HERE: Instead of a string I want an error object.
error = v8::String::New( m_errorMsg.c_str() );
}
v8::Handle< v8::Value > argv[] = { error, result };
m_callback->Call(v8::Context::GetCurrent()->Global(), 2, argv);
actually the api was changed. now, you can throw an exception in this way
...
Isolate* isolate = args.GetIsolate();
isolate->ThrowException(Exception::TypeError(
String::NewFromUtf8(isolate, "Your error message")));
...
This is not specific to Node, you can just use the ThrowException method from the V8 API. It requires one parameter of type Exception which you can create with one of the static methods on the Exception class.
There are examples on how to do it in this Node tutorial but feel free to check out my code on GitHub too.
ThrowException(Exception::TypeError(String::New("This is an error, oh yes.")));
return Undefined();
NOTE: Don't forget to return with Undefined() after calling ThrowException. (Even scope.close is unnecessary in this case.)
Further reading:
documentation of the ThrowException method:
http://bespin.cz/~ondras/html/namespacev8.html#a2469af0ac719d39f77f20cf68dd9200e
documentation of the Exception class:
http://bespin.cz/~ondras/html/classv8_1_1Exception.html
As far as i know there isn't a v8::Error class but there's the v8::Exception that provides static member functions to construct different types of errors by supplying a v8::String argument.
Take a look at the v8::Exception Class Reference, that's probably what you're looking for.
I hope this helps.
Imaging a class which is doing the following thing
class AClass
{
AClass() : mode(0) {}
void a()
{
if (mode != 0) throw ("Error mode should be 0");
// we pass the test, so now do something
...
mode = 1;
}
void b()
{
if (mode != 1) throw("Error mode should be 1");
// we pass the test, so now do something
...
}
int mode;
};
The class contains many methods (easily than 20) and for each one of these methods we need to do a check on the value of mode which is obviously a lot of code duplication. Furthermore, we can identify two categories of methods, those who will throw an error if mode !=0 and those who will throw an error if mode != 1. Could it somehow be possible to group these methods in two categories (category A = method who throw an error if mode != 0) and category B for method who throw an error if mode != 1)?
EDIT: Looking at the current answers I realise the way I formulate the question and the problem is probably not clear enough. What I want to avoid is to have to call for a function in each method of the class. Whether we write code at the beginning of the methods or put this code in a function and call this function is not the problem. The question is whether we can avoid this all together. Whether there is a technique that would help to automatically check whether the call to a method of a class is valid depending on some context.
AClass is actually an API in the context of my project. a(), b(), etc. are some functions that the programmer can call if she/he wants to use the API however some of these methods can only be called in some precise order. For example you can see in the code that a() sets mode = 1. So the programmer could do something like this:
a(); // mode = 0 so it's good
b(); // mode = 1 so it's good
but this code needs to fail (it will compile of course but at execution time I need to throw an error mentioning that the context in which b() was called was wrong.
b(); // mode 0 so it won't work
a(); // it will compile but throw an exception
I tried to see if any pattern could work for doing this but couldn't find anything at all. It seems impossible to me and I believe the only option is really to write the necessary code. Could anyone though suggest something? Thank you very much.
Just add private member functions:
void assert_mode_0() {
assert_mode(0);
}
void assert_mode_1() {
assert_mode(1);
}
void assert_mode(int m) {
if (mode != m)
throw msg[m];
}
with a suitable definition of msg, of course.
Aside from implementing the check in a dedicated method (a great suggestion), you could also consider decomposing the behavior in AClass into two distinct classes, or delegate the specific portion to a new pair of classes. This seems especially appropriate if the mode is invariant for an instance (as it is in the example).
Well I guess the simplest solution would be defining a macro or some inline function like this:
#define checkErrorMode0(x) \
if ((x) != 0) throw ("Error mode should be 0");
#define checkErrorMode1(x) \
if ((x) != 1) throw ("Error mode should be 1");
// or, probably within your class
inline void checkErrorMode0(int x){
if ( x != 0 ) throw ("Error mode should be 0");
}
inline void checkErrorMode1(int x){
if ( x != 1 ) throw ("Error mode should be 1");
}
So you could simply call one of these methods inside of the functions that require them.
But most likely there is a more elegant workaround for what you want to do.
After looking into the problem a bit more, it seems that the closest helpful answer is (by Nick):
Try looking into Aspect Oriented Software Development en.wikipedia.org/wiki/Aspect-oriented_software_development – Nick
The Wikipedia page is not easy to read and doesn't provide a C++ example, so it stays very abstract at first, but if you search for Aspect Oriented Programming and C++ you will find links with examples.
The idea behind it (and it just a very quick summary) is to find a way of adding "services" or "functionalities" to a class. These services can notably be added at compile time through the use of templates. This is what I was intuitively experimenting with as an attempt at solving my problem, and I am glad to see this technique has been around for many years.
This document is a good reference:
Aspect-Oriented Programming & C++ By Christopher Diggins, August 01, 2004.
And I found this link with example useful to understand the concept:
Implementing Aspects using Generative Programming by Calum Grant.
I have written a few C++ classes which employ a variety C++ libraries. I made a Windows Form project, and set it up to use my classes successfully. However, I recently made another C++ class and now I consistently get:
A first chance exception of type 'System.AccessViolationException' occurred in TEST_OCU.exe
which leads to:
An unhandled exception of type 'System.TypeInitializationException' occurred in Unknown Module.
Additional information: The type initializer for '<Module>' threw an exception.
The program hasn't even started running yet, and the new, problem-causing C++ class hasn't even been constructed yet. If I comment out the new call, and only have a pointer to this new C++ class, everything compiles just fine. BUT, if somewhere I do something like:
if(new_class_ptr != NULL)
new_class_ptr->SomeFunction() //It doesn't matter what function this is
This will throw those violations again
Some facts:
Compiling and linking is fine, this seems to be a run-time problem.
My solution employs unmanaged C++ libraries and classes (that I have written), and one managed C++ Form.
So far I haven't had any problems, I've used a few C++ libraries successfully for a while. This is caused by a new C++ class I recently wrote.
The C++ class which causes these violations uses std::ifstream to read in a file. If I comment out the std::ifstream input_file(filename); my Forms project runs successfully.
If I use the C++ class in a simple Win32 project, it compiles and runs just fine with the std::ifstream.
I have a strong feeling it is related to this question
Could anyone offer any advice? Thank you
EDIT: I'm providing some parts of my form code I have. RTSPConnection works just fine, the offending class is RTPStream
public ref class Form1 : public System::Windows::Forms::Form
{
public:
// ... Lots of generated code here ...
//Calls I've written
private: static RTSPConnection * rtsp_connection_ = NULL; //This class works
private: static RTPStream * rtp_connection_ = NULL; //This class does not
bool streaming_;
System::Threading::Thread^ streaming_thread_;
private: System::Void Form1_Load(System::Object^ sender, System::EventArgs^ e) {
if(rtsp_connection_ == NULL)
{
rtsp_connection_ = new RTSPConnection("rtsp://192.168.40.131:8554/smpte");
streaming_ = false;
}
//if(rtp_connection_ == NULL)
// rtp_connection_ = new RTPStream("test");
}
private: System::Void Form1_FormClosing(System::Object^ sender, System::Windows::Forms::FormClosingEventArgs^ e) {
if(rtsp_connection_ != NULL)
rtsp_connection_->StopStreaming();
}
private: System::Void button1_MouseClick(System::Object^ sender, System::Windows::Forms::MouseEventArgs^ e) {
if(!streaming_)
{
//Start Streaming
streaming_thread_ = gcnew Thread(gcnew ThreadStart(&Form1::WorkerThreadFunc));
streaming_thread_->Start();
this->button1->Text = L"Stop Streaming";
streaming_ = true;
}
else
{
//Stop Streaming
if(rtsp_connection_ != NULL)
rtsp_connection_->StopStreaming();
//THIS CALL RIGHT HERE WILL THROW AN ACCESS VIOLATION
if(rtp_connection_ != NULL)
rtp_connection_->StopStreaming();
this->button1->Text = L"Start Streaming";
streaming_ = false;
}
}
};
These two statements appear to contradict each other:
The program hasn't even started running yet, and the new,
problem-causing C++ class hasn't even been constructed yet.
If I comment out the new call, and only have a pointer to this new C++
class, everything compiles just fine.
Q: Could you please post the code where you're calling "new"? Or are you calling "new" - perhaps you just meant "if I comment out my new class"?
Q: Could you please set a breakpoint in your constructor, look at the stack trace, and see who's invoking it? And when?
========== ADDENDUM ==========
I strongly disagree with this statement:
It all hinges on this line: std::ifstream input_file(filename);
where filename is a std::string.
I strong AGREE with this statement:
You get almost the same error in C# if you have static class members
that depend on each other and they aren't initialised in the order you
expect. In C++, if you had a static singleton and another static
member that referred to it
Calling "ifstream" isn't the problem per se. Rather, somehow invoking the class that invokes ifstream before the program has initialized is the problem.
Q: Are you calling "new" on this class? If so, where. Please cut/paste that code.
Per MSDN, you should be able to set "mixed mode debugging". I have lots of different copies of MSVS :) ... but MSVS 2010/C++ doesn't happen to be one of them. Please look at this documentation:
http://msdn.microsoft.com/en-us/library/fz5w87ad.aspx
http://msdn.microsoft.com/en-us/library/cktt23yw