Re: using namespace in xcode and dev-c++
Re: using namespace in xcode and dev-c++
- Subject: Re: using namespace in xcode and dev-c++
- From: Jeff DuMonthier <email@hidden>
- Date: Thu, 30 Mar 2006 20:25:18 -0500
I've never seen this usage and it seems very strange as well as
completely at odds with what a namespace is.
The usual technique for including files only once is:
[in the header file]
#ifndef GLOBALSTUFF_H
#define GLOBALSTUFF_H
...
(stuff to include)
...
#endif
[in the .cpp file]
#include "globalstuff.h"
...
(source)
...
Some compilers provide nonstandard alternatives to #include which
prevent the same file being included more than once and eliminate the
need for the #ifndef, #define and #endif in the header. In objective
C/C++, #import is a standard directive which does just that.
A namespace is a different animal. When you have a structure like this:
namespace globalstuff
{
...
}
Anything declared inside will have globalstuff:: prepended to its name.
To reference a function f declared in the namespace, you would have to
use globalstuff::f(), unless you also had:
using namespace globalstuff;
In that case everything in the namespace globalstuff is brought out and
made accessible in the current namespace without explicitly specifying
it. You might still have to if some names are hidden by other names
already visible in the namespace, or something like that. The function
of a namespace is not to include headers only once, it is to avoid name
collisions.
Maybe some compilers you used had some 'smart' feature which guessed
that you really wanted namespace globalstuff and just ignored the '.h'.
I can't say, but the name of the namespace should be the same in the
using directive and the namespace declaration or it really doesn't make
sense. Also, a namespace really should not prevent the reading of a
header file multiple times. I might even guess that it wasn't before
and some combination of multiple reading and nonstandard interpretation
somehow caused the code to work when it shouldn't have.
On Mar 30, 2006, at 10:23 AM, Jeffery Shipman wrote:
I am trying to hammer out a couple differences between compilers. I
use xcode and dev-c++. The use of namespace is different between the
two compilers.
In Dev-C++ you may put the following in a .h or .cpp file
#include "globalstuff.h"
using namespace globalstuff.h;
In the associated header is the following
#include <cstdlib>
#include <iostream>...
namspace globalstuff
{
#include "node.h"
#include "node.cpp" ..
}
This avoids duplication of headers and libraries. This works fine in
Dev-C++ but produces errors in xcode (The .cpp file does not see it's
associated .h file). My readings in my class textbooks suggests this
should work. Does anyone know why it does not work in xcode.
Thanks
Jeffery Shipman
email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Xcode-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
email@hidden
This email sent to email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Xcode-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden