Chapter 14: Error Handling
Historically, managing runtime errors has always been a hard problem to solve because of their complex and different natures, spanning from hardware failures to business logic errors.
Some of these errors, such as division by zero and null dereferencing, are generated by the CPU itself as an exception, while others are generated at the software level and propagated either as an exception or as an error code, depending on the runtime and programming language.
The .NET platform has been designed to manage an error condition through an exception strategy, which has the big advantage of dramatically simplifying the handling code. This means that any property or method may throw an exception and communicate the error condition through exception objects.
Throwing exceptions raises an important question—is the exception part of the contract between the library implementor and its consumer, or is it, rather, an implementation detail?
In this chapter...
