RE: dealloc (Correction)

Lim Hock-Chai wrote on 07/12/2006 02:09:02 PM:

Good point, Scott.  From everybody's posting, seems like there is really
no reason to ever set based pointer to *null by assigning *null to it.

Sure there is.  If I call a subprocedure that is supposed to return a 
pointer to an object, it should probably return *null if it can't find the 
object for some reason.  If I wrote such a subprocedure, I'd have a strong 
argument for setting a pointer to *null.

In fact, pointers which point to  deallocated memory should probably be 
set to null all the time (as Barbara said early, IIRC).  That way, you 
*always* get an error if you try to reuse that pointer later in the code 
without resetting it to point to valid memory, instead of just maybe 
getting an error.

#####################################################################################
Attention:
The above message and/or attachment(s) is private and confidential and is 
intended 
only for the people for which it is addressed. If you are not named in the 
address 
fields, ignore the contents and delete all the material. Thank you. Have a nice 
day.

For more information on email virus scanning, security and content
management, please contact administrator@xxxxxxxxxxxx
#####################################################################################