Catching Exceptions in C : By Value or By Reference?

Exception Handling in C : Catching by Reference vs. Value

Catching exceptions by value has been a common practice in C . However, the standard recommendation suggests a different approach.

Best Practice: Throw by Value, Catch by Reference

The recommended practice is to throw exceptions by value and catch them by reference. This is known as "copy-on-write" semantics.

Advantages of Catching by Reference

Catching exceptions by reference offers several benefits:

• Preserves Instance: The caught exception retains its original type and state, ensuring that inherited functionality is not affected.
• No Unnecessary Copies: Since the exception is passed by reference, no additional copies are created, improving performance and reducing memory usage.

Example:

Consider this example:

class CustomException { int errorCode; };
class MyException : public CustomException { int customCode; };

try {
  // Code that potentially throws an exception
}
catch (CustomException& e) {
  // Handle the exception, preserving its type and state
}

When to Consider Catching by Value

While catching by reference is generally recommended, there are exceptional cases where catching by value may be appropriate:

• Error Reporting: If the goal is simply to report an error and not take action based on the exception's contents, catching by value can simplify handling.
• Saving Execution Space: In very limited scenarios where stack space is extremely low, catching by value can save some bytes on the stack frame.

The above is the detailed content of Catching Exceptions in C : By Value or By Reference?. For more information, please follow other related articles on the PHP Chinese website!