[fdo-dev] Checking new'ed pointers for NULL

[Traian Stanev]

The delete operator is a noop for NULL pointers, according to the C++
standard.

-----Original Message-----
From: Greg Boone 
Sent: Monday, August 28, 2006 11:43 AM
To: dev at fdo.osgeo.org
Subject: RE: [fdo-dev] Checking new'ed pointers for NULL

Why do you advocate not to test pointers for NULL before calling delete
operator?

-----Original Message-----
From: Mateusz Loskot [mailto:mateusz at loskot.net]
Sent: Monday, August 28, 2006 11:09 AM
To: dev at fdo.osgeo.org
Subject: Re: [fdo-dev] Checking new'ed pointers for NULL

Pierre Dalcourt wrote:
> Hi Mateusz,
> 
> I think most of us agree with you.  The code that checks for a new 
> operator returning NULL is either old code or code that was written by

> developers (including myself) before they realized that the new
operator
> on modern compilers would never actually return NULL.  :-)  However, 
> code that does the extra check is usually not harmful in itself so it 
> may not be worth spending the extra effort to remove them unless you 
> happen to be working on those files...


I see and I agree, though for me this redundant tests make it harder to
read and analyze the code, especially when function bodies are >100
lines long.

So, to summary this thread:
- it's OK to to not to test pointers for NULL after allocation with new
operatir
- it's OK to not to test pointers for NULL before calling delete
operator.

If there are no objections, I'm following those rules in my code for
FDO.

Cheers
--
Mateusz Loskot
http://mateusz.loskot.net

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe at fdo.osgeo.org For additional
commands, e-mail: dev-help at fdo.osgeo.org



---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe at fdo.osgeo.org For additional
commands, e-mail: dev-help at fdo.osgeo.org