Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

NSAssert vs. assert: Which do you use, and when?

I've read two really interesting pieces of advice, recently:

  1. In the comments to this StackOverflow answer, @Mike Weller says to leave your asserts on in production code... what's the performance hit, really? Is there any reason NOT to leave them in?
  2. In Vincent Gable's blog, he states that you should prefer assert over NSAssert... is there any reason NOT to use assert? (it's less letters :))
like image 745
Dan Rosenstark Avatar asked Jul 07 '11 19:07

Dan Rosenstark


People also ask

When should I use assert?

assert() macro is used to test the conditions or assumptions that should not occur in a program. For example, the array index should always be > 0. Another assumption can be 2+2 == 3+1. So using assert () we can test such assumptions and as long as they evaluate to true, our program runs normally.

Why use assert instead of if in python?

That an “Assert” is used only for validations, where an “If” clause is used for the logic within our code. We can use an “If” clause to determine whether our automation should follow one path or another, but an “Assert” statement to validate the elements within those paths.

What is NSAssert?

The NSAssert macro evaluates the condition and serves as a front end to the assertion handler. Each thread has its own assertion handler, which is an object of class NSAssertionHandler . When invoked, an assertion handler prints an error message that includes the method and class names (or the function name).

Why we use assert function?

The assert() function tests the condition parameter. If it is false, it prints a message to standard error, using the string parameter to describe the failed condition. It then sets the variable _assert_exit to one and executes the exit statement. The exit statement jumps to the END rule.


1 Answers

To answer your two questions:

  1. There should be very little performance hit for leaving in an assertion, unless the actual action in the assertion is time consuming (e.g. assert([obj calculateMeaningOfLife] == 42)). An assertion should be no different than an extra if statement, performance-wise. The reason for stripping out assertions in release builds is that they are essentially a debugging tool--they catch inconsistent internal program state at runtime. From a developer perspective, it's much better for an app to crash as soon as something goes wrong, but from a user perspective it's arguably less annoying if the app doesn't crash (unless letting the app run with abnormal state causes something horrible to happen), and exposing development details in error messages can be off-putting. There are good arguments on both sides--if I remember correctly, Code Complete recommends stripping them out but The Pragmatic Programmer recommends leaving them in. In any case, assertions are not a substitute for proper error handling and should only be used for programming errors.

  2. The basic difference between an NSAssert and a regular assert is that an NSAssert raises an exception when it fails while an assert just crashes the app. NSAssert also lets you supply fancier error messages and logs them. Practically, I really don't think there's much difference between the two--I can't think of a reason to handle an exception thrown by an assertion. (To split hairs, I think NSAssert usually involves less typing because you don't have to include assert.h, but that's neither here nor there.)

like image 88
shosti Avatar answered Sep 21 '22 05:09

shosti