Mailing Lists: Apple Mailing Lists

Image of Mac OS face in stamp
 
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: NS_HANDLER mystery



On Jul 29, 2005, at 5:53 PM, Will Mason wrote:
Why does _NSRemoveHandler() never get called upon reaching
NS_ENDHANDLER? To me the logic would dictate that the structure should
either always be removed from the internal collection, or never. Having
three cases where it's removed and one case where it isn't just doesn't
make any sense to me.


Can anyone enlighten me?

Look closely at the macros:

#define NS_DURING { NSHandler2 _localHandler;            \
            _NSAddHandler2(&_localHandler);        \
            if (!_NSSETJMP(_localHandler._state, 0)) {

#define NS_HANDLER _NSRemoveHandler2(&_localHandler); } else { \
NSException *localException = _NSExceptionObjectFromHandler2(&_localHandler);


#define NS_ENDHANDLER localException = nil; /* to avoid compiler warning */}}

NS_DURING creates *two* scopes; one created by the initial { and the other created by the { of the IF statement. There is a third scope created by NS_HANDLER which corresponds to the path of execution of the else clause of the if statement.

Now, go read the _setjmp man page.  It specifically says:

The sigsetjmp(), setjmp(), and _setjmp() functions save their calling
environment in env. Each of these functions returns 0.


The corresponding longjmp() functions restore the environment saved by
their most recent respective invocations of the setjmp() function. They
then return so that program execution continues as if the corresponding
invocation of the setjmp() call had just returned the value specified by
val, instead of 0.


In other words, the call _NSSETJMP (which really just calls _setjmp) returns 0 and the if() statement then enters the body of the second scope defined by NS_DURING. This scope is the execution path prior to an exception being raised.

Raising an exception invokes longjmp(), which causes execution to effectively GOTO the evaluation of the if() statement, only this time _NSSETJMP() will return a non-zero value, thus causing the else clause of the NS_DURING's if statment to be executed.

During execution of the else code path, the exception catching scope has already been destroyed. So, no reason to remove a handler that has already been removed as a part of raising an exception.

When an exception is raised, the exception handler is consumed because that particular handler can never be used again as there is no way for control flow to return to the NS_DURING/NS_HANDLER code path. Thus, no need to remove the handler and, as a result, no need to use the NS_*RETURN() functions within the NS_HANDLER/NS_ENDHANDLER blocks.

b.bum

_______________________________________________
Do not post admin requests to the list. They will be ignored.
Cocoa-dev mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:
http://lists.apple.com/mailman/options/cocoa-dev/email@hidden

This email sent to email@hidden
References: 
 >NS_HANDLER mystery (From: Will Mason <email@hidden>)



Visit the Apple Store online or at retail locations.
1-800-MY-APPLE

Contact Apple | Terms of Use | Privacy Policy

Copyright © 2007 Apple Inc. All rights reserved.